Scientific Article

Studies in Mechanobiology, Tissue Engineering
and Biomaterials
Volume 5
Series Editor
Amit Gefen, Ramat Aviv, Israel
For further volumes:

http://www.springer.com/series/8415
Matthew J. Silva
Editor
Skeletal Aging
and Osteoporosis
Biomechanics and Mechanobiology
123
Editor
Matthew J. Silva
Department of Orthopaedic Surgery
Washington University, St. Louis
MO, USA
ISSN 1868-2006 ISSN 1868-2014 (electronic)

ISBN 978-3-642-18052-1 ISBN 978-3-642-18053-8 (eBook)
DOI 10.1007/978-3-642-18053-8
Springer Heidelberg New York Dordrecht London
Library of Congress Control Number: 2012943364
Ó Springer-Verlag Berlin Heidelberg 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief
excerpts in connection with reviews or scholarly analysis or material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the
work. Duplication of this publication or parts thereof is permitted only under the provisions of
the Copyright Law of the Publisher’s location, in its current version, and permission for use must always
be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright
Clearance Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Preface
Why another bone book? I agreed to edit this book because there is no similar book
that I know. There are excellent texts covering bone mechanics (e.g., by Cowin) and
musculoskeletal biomechanics (e.g., by Bartel, Davy and Keaveny; Martin, Burr and
Sharkey; Mow and Huiskes), and equally excellent (and massive) texts covering
bone biology/aging/osteoporosis (e.g., by Marcus, Feldman, Nelson and Rosen;
Rosen, Glowacki and Bilezikian). In these texts, the topic of bone biomechanics and
aging is just a small part of a larger agenda. Here my goal was to narrow the focus
and devote an entire volume to the questions: What changes in bone(s) occur with
aging or osteoporosis that are relevant to bone strength? How do we predict bone
strength? How do osteoporosis drugs affect bone strength? What changes occur with
aging that are relevant to bone mechanobiology? There has been a lot of research on
these questions in the past 40 years, but no single volume that attempts to review it.
The assembled chapters offer such a review. They highlight many age-related
phenomena that are irrefutable, but also point to issues that are debatable or not fully
explored. Aging studies are difficult whether they use animals, human subjects or
post mortem material, and there is still much work to be done.
The first five chapters address the biomechanics question. Chapter 1 covers
changes in bone structure and strength at the whole-bone level, while Chaps. 2–5
cover changes in properties at the trabecular and cortical bone tissue level, with
focus on microstructure, composition and microdamage. Chapter 6 reviews recent
attempts at integrating our knowledge of structure, strength and loading to predict
fracture risk. Chapter 7 reviews the effects of osteoporosis drug treatments on bone
strength and fracture. The next four chapters address the mechanobiology
question. Chapter 8 reviews mechanoresponsiveness and aging at the cellular
level. Chapters 9 and 10 review mechanoresponsiveness in animal experiment,
with focus on aging and sex hormones, respectively. Lastly, Chap. 11 reviews
clinical evidence that loading influences bone in the setting of aging/osteoporosis.
Even a modest volume like this takes a large collective effort. I heartily thank
each of the authors who contributed chapters to this volume. They generously gave
of their time to write and revise their chapters. I hope that readers will find our
efforts were worthwhile.
v
Contents
Age-Related Changes in Whole-Bone Structure and Strength . . . . . . . 1

Matthew J. Silva and Karl J. Jepsen
Characterisation of Trabecular Bone Structure . . . . . . . . . . . . . . . . . 31

Ian H. Parkinson and Nicola L. Fazzalari
Cortical Bone Mechanics and Composition:

Effects of Age and Gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Xiaodu Wang
Bone Microdamage and Its Contributions to Fracture . . . . . . . . . . . . 87

Lamya Karim and Deepak Vashishth
Changes in Cortical Bone Mineral and Microstructure

with Aging and Osteoporosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Janardhan Yerramshetty and Ozan Akkus
Factor of Risk for Fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Dennis E. Anderson and Mary L. Bouxsein
Bisphosphonates and PTH for Preventing Fractures . . . . . . . . . . . . . . 151

David B. Burr and Matthew R. Allen
Bone Cell Mechanoresponsiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Damian C. Genetos and Christopher R. Jacobs
vii
viii Contents
The Effect of Aging on Skeletal Mechanoresponsiveness:

Animal Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Akhilesh A. Kotiya and Matthew J. Silva
Skeletal Mechanoresponsiveness: Effects of Sex Hormones . . . . . . . . . 217

Katherine M. Melville, Natalie H. Kelly
and Marjolein C. H. van der Meulen
Effects of Exercise and Physical Interventions on Bone:

Clinical Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Wendy M. Kohrt, Karen L. Villalon and Daniel W. Barry
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257

Age-Related Changes in Whole-Bone
Structure and Strength
Matthew J. Silva and Karl J. Jepsen
Abstract We review data on age-related changes in bone geometry of relevance

to whole-bone strength, as well as the limited data on changes in strength. Con-
sistently across many sites, women have bones that are smaller (by 15–30 %) than
age-matched men, and thus are weaker. In both women and men, modest periosteal
expansion of the diaphysis occurs throughout life, but this is accompanied by a
faster rate of medullary expansion, especially in women. The net result is an age-
related decrease in cortical bone at most sites in women, but negligible changes in
men. At metaphyseal sites there is also modest periosteal expansion as well as
endosteal expansion and net cortical bone loss. But the dominant change with
aging is decreased trabecular bone density, with most studies showing greater rates
of decline in women than men. These effects are especially pronounced at the
proximal femur and vertebra. Changes in whole-bone strength with aging are less
well documented. Available data (from mechanical tests and computer models)
suggest modest declines in diaphyseal strength in women but not men, and much
greater declines in strength of the proximal femur and vertebra. Women and men
appear to lose proximal femur strength at similar rates, although the decline starts
earlier in women. Also, both women and men lose vertebral strength with aging,
with some data indicating a faster decline in women but other indicating
M. J. Silva (&)
Department of Orthopaedic Surgery, Washington University School of Medicine,
Saint Louis, Missouri 63110, USA
e-mail: silvam@wustl.edu
K. J. Jepsen
Department of Orthopaedic Surgery, University of Michigan Ann Arbor,
Michigan 48109, USA
e-mail: kjepsen@umich.edu
Stud Mechanobiol Tissue Eng Biomater (2013) 5: 1–30 1

DOI: 10.1007/8415_2012_137
Published Online: 13 June 2012
2 M. J. Silva and K. J. Jepsen
equivalent rates of decline. In conclusion, there are important age-related changes

in bone structure and density that affect whole-bone strength. Additional studies
measuring whole-bone strength with aging are needed.
1 Introduction
Bones are structural entities that are increasingly susceptible to fracture with aging.
Age-related/osteoporotic fractures occur at an estimated rate of 2 million per year in
the U.S. with corresponding high costs in economic terms, quality of life, and
increased mortality [1]. The causes of the increase in fracture incidence with age are
multifactorial, but can generally be grouped into factors affecting applied loading
(e.g., body weight, impact from falls or other trauma) and factors affecting structural
(whole-bone) strength. Because the mechanical behavior of a structure depends on
its geometric and its material properties, changes in geometry and material prop-
erties of bones with age influence whole-bone strength. The changes that occur in the
material properties of bone with age include density, microstructure, composition,
etc., and are considered in other chapters in this volume (‘‘Characterisation of
Trabecular Bone Structure, Cortical Bone Mechanics and Composition: Effects of
Age and Gender, Bone Microdamage and its Contributions to Fracture, Changes in
Cortical Bone Mineral and Microstructure with Aging and Osteoporosis’’, ‘‘Bone
Microdamage and its Contributions to Fracture’’, Changes in Cortical Bone Mineral
and Microstructure with Aging and Osteoporosis’’). Of primary interest in this
chapter are the changes in bone structure (i.e., size and shape, also called mor-
phology) that are documented to occur with aging. We also consider the limited
available data on whole-bone mechanical properties and aging.
There are many descriptors of bone morphology, but based on engineering
mechanics we focus on two geometric properties of particular relevance to whole-
bone strength: cross-sectional area and moment of inertia. For example, for a
cylindrical structure like the diaphysis (shaft) of a long bone (Fig. 1), the theo-
retical strength under axial and bending loading are given by:
Ffail ¼ rfail x A
Mfail ¼ rfail x I = c
where Ffail is the axial failure force (structural strength as it pertains to failure
under purely compressive or tensile loads), Mfail is the bending failure moment
(structural strength as it pertains to failure under bending loads), rfail is the failure
stress (material strength), A is the cross-sectional area, I is the cross-sectional
moment of inertia (also called the second moment of area), and c is the distance
from the center of the cross-section to the outer most point on the surface. For a
solid cylinder with a circular cross-section: A = pD2/4, I = pD4/64, and c = D/2,
where D is the diameter. (See Fig. 1 for additional equations.) From these two
Age-Related Changes 3
Fig. 1 Sketch of idealized a diaphyseal and b metaphyseal cross-sections of bones. For the
hollow circular cross-section of a: total area, TA = pD2P/4; medullary area, MA = pD2M/4;
cortical bone area, CA = TA-MA = p(D2P-D2M)/4; bone moment of inertia = p(D4P-D4M)/64
equations we note that structural strength depends directly on rfail (material) and
either A or I (geometry), which in turn depend on diameter squared or raised to the
power four. Area is essentially a measure of the amount of bone in the cross-
section, while moment of inertia is a measure reflecting both the amount of bone
and how it is distributed. These simple relationships motivate our interest in area
and moment of inertia as two key measures of bone size. Also of interest are
diameter and section modulus, defined as I/c.
Bones are dynamic structures that change throughout life. In this chapter, we
review the published data on changes in bone structure with aging. We consider
changes in the diaphyseal regions of long bones (e.g., femur, tibia, radius), which
are comprised mostly of cortical bone, and changes at the ends of long bones and
in short bones (e.g., vertebra), which are comprised of a mix of cortical and
trabecular bone. These latter sites are of particular clinical relevance as the most
common fracture sites are the vertebra, distal radius and proximal femur. We focus
primarily on bone geometry, but also consider whole-bone mechanical properties
where data are available. We also discuss recent work showing that material and
geometric properties, often considered to be independent contributors to whole-
bone strength, are probably not independent.
2 Diaphyseal Changes with Age
2.1 Cross-Sectional Geometry
Even after rapid skeletal growth ends in the third decade of life, the diaphyses of
long bones continue to change via periosteal and endocortical expansion. The net
changes in geometry indicate that bone apposition prevails at the periosteum,
while resorption prevails at the endocortex. Generally, these changes occur

throughout life, although the rates of expansion are site- and sex-dependent. Here
we review changes with age in the diaphyseal cross-sections of bones, specifically
subperiosteal area (also called total area, TA), medullary area (MA) and cortical
bone area (CA).
2.1.1 Lower Extremity: Femur and Tibia
The size of the femoral and tibial diaphysis in women is approx. 20 % smaller than
age-matched men. This figure is an average across the studies we reviewed [2–9],
and refers to both total area and cortical area.
The phenomenon of diaphyseal expansion of weight-bearing long bones with
aging has been recognized for at least 50 years. Early studies were based on
femoral radiographs [10] or physical sections from cadaver bones [2, 4, 5]. In a
well-known ‘‘classic’’ study, Ruff and Hayes [5] reported that women do not
exhibit age-related periosteal expansion as men do, although this result was based
on only 37 female donors, and later studies have contradicted this conclusion.
More recently, non-invasive imaging methods such as peripheral quantitative
computed tomography (pQCT) have been used to determine true cross-sectional
geometry in large numbers of subjects [3, 6–9].
The findings of several classic and recent studies are summarized in Table 1,
where age-related changes are presented as average percent change per 10 years of
life based on linear regression analysis. These results support a consistent pattern.
In women, the periosteum expands at a slow rate from young adulthood (3rd
decade) to old age, resulting in an average increase in total area of *2 % per
decade. However, the endocortex expands at a faster rate, resulting in an average
increase in medullary area (MA) of *13 %/decade and a net loss of cortical bone
area of *3 %/decade. In men, the rate of periosteal expansion is similar as in
women (*2 %/decade), but the medullary expansion is notably slower (*7 %/
decade) and the net change in cortical area is negligible (-1 %/decade). Because
moment of inertia is related to the fourth power of bone diameter, the age-changes
in moment of inertia reflect primarily the age-changes in total area and to a lesser
but important extent the age-changes in cortical area. Both parameters are needed
to fully understand the impact of periosteal expansion and endocortical resorption
on the resistance to bending and torsional loading. Data on age-changes in moment
of inertia (or section modulus) are limited. The majority of reports indicate no
significant age-effect on moment of inertia, due in part to statistically under-
powered studies (i.e., type II error). Of the studies using 3D imaging methods with
large sample sizes, Russo et al. [6] reported a decline in density-weighted moment
of inertia in women but no change in men, while Yuen et al. [9] reported a slight
decline (-2 %/decade) in both women and men. A decline with aging might be
interpreted as a failure of biological processes to maintain mechanical function
throughout life.
Table 1 Diaphyseal changes—lower extremity
% Change per decade*
Authors Materials Method Study Bone Sex D_P D_M TA MA CA Moment
design of inertia
a
Smith & Walker [10] 2030 women 45–90 years AP radiographs CS Femur Female 3.1 10.2 6.4 b 21.3 b 2.9 b 8.6 b, c
Martin & Atkinson [4] 31 cadavers (18 female, Bone sections CS Femur Female -1.1 (NS) 0.8 (NS) -2.1 b (NS) 1.7 b (NS) -5.0 -5.0 (NS)
13 male) 22–82 years d Male 1.0 (NS) 3.1 (NS) 2.1 b (NS) 6.2 b (NS) -3.7 (NS) 2.8 (NS)
Age-Related Changes
a
Feik et al. [2] 180 cadavers (87 female, Bone sections CS Femur Female – – 4.1 17.9 -2.7 –
93 male) 21–100 years Male – – 2.6 10.3 -0.2 (NS) –
Sigurdsson et al. [8] 1715 subjects (908 female, QCT CS Femur Female – – 3.1 e 14.7 f -3.7 e, g –
807 male) 67–93 years Male – – 2.8 e 4.9 f 0.4 e, g –
Ruff & Hayes [5] 75 cadavers (37 female, 38 Bone sections CS Femur Female – – 1.1 (NS) 10.8 -5.9 -1.8 (NS)
male) 20–99 years Male – – 1.8 8 -1.6 (NS) 1.9 (NS)
Tibia Female – – 1.0 (NS) 10.4 -5.6 -2.0 (NS)
Male – – 2.7 8.8 -0.1 (NS) 3.7
a
Russo et al. [6, 7]h 1205 subjects (693 female, pQCT CS Tibia Female – – 0.6 11.6 -3.1 -1.6 i
512 male) 20–102 years Male – – 3.1 9.9 0.5 (NS) -0.6 (NS) i
(continued)
5
6
Table 1 continued
Authors Materials Method Study design Bone Sex D_P D_M TA MA CA Moment of inertia
e j
Lauretani [3] 809 subjects (464 female, pQCT CS, LNG Tibia Female – – 0.8 10.4 -2.1 –
345 male) 21–102 years Male – – 1.2 3.7 0.6 –
a, k
Yuen [9] 1258 subjects (638 female, pQCT CS Tibia Female – – – – -2.7 -1.9
620 male) 20–98 years Male – – – – -1.4 -1.8
D_P, periosteal diameter; D_M, medullary diameter; TA, total (subperiosteal) area; MA, medullary area; CA, cortical bone area
All results based on cross-sectional (CS) study design, except Lauretani which is mix of cross-sectional and longitudinal (LNG)
*Percent change calculated using slope of regression line (e.g, TA vs. age), multiplied by 10 and divided by overall mean (unless noted otherwise)
Slopes are taken from papers (if reported), or determined by linear regression from data presented in papers
(NS) regression is not statistically significant, p [ 0.05
Femur: 50 % site (mid-shaft); Tibia: approx. 35 % site (mid-distal); section location is given as percent of limb length, relative to the distal end
a
regressions based on mean values for the age groups reported
b
authors computed areas assuming a circular cross-section
c
section modulus was reported
d
study included six subjects \20 yrs age; these are excluded here
e
from Table 3 of paper
f
estimated from Fig. 2 of paper
g
cortical thickness index is reported, not cortical area
h
Russo (2003, 2006) and Lauretani (2008) report data from same study (InCHIANTI), but Laurentani includes longitudinal data
i
density-weighted moment of inertia reported
j
from Table 4 of paper, recalculated based on estd. overall means
k
Yuen reports slopes for \60 and [60 years in Table 4 of paper; values shown here are computed from slopes across all age groups
M. J. Silva and K. J. Jepsen
It is important to note that most studies have been cross-sectional in design, i.e.,
have inferred changes with aging based on differences between groups of donors/
subjects of different ages. This approach has limitations due to secular trends such
as increased stature over the past century. Thus, an older age group is likely to be
shorter than a younger age group, which will introduce bias because of the known
dependence of bone cross-sectional size on bone length [5]. One way to account
for this is to scale the cross-sectional properties by length, which is done in some
but not all studies.
The ideal study design for assessing bone changes with aging is to follow the
same individuals over time (longitudinal). In a longitudinal study, Lauretani et al.
[3] performed pQCT scans of the distal tibia in 809 subjects (females and males) at
three timepoints (0, 3, 6 years follow-up). They noted that the age-related changes
in bone parameters were underestimated by cross-sectional analysis compared to
the ‘‘true’’ rates of change measured longitudinally. (This is consistent with the
trend that Garn et al. reported for metacarpals [11]; see Table 2.) In addition, the
rates of change vary with age and there are age-sex interactions. This is clearly
illustrated in data from their study (Fig. 2). For example, both women and men
show an overall increase in subperiosteal area (‘‘total bone area’’) with aging, but
they show opposite trends for rates of change. Women have little change early in
life but steep increases later in life ([70 years), while men have relatively steep
increases early (20–40 years) which slow to zero and then actually reverse with
old age ([80 years). While these particular findings may be unique to the tibial site
and the study population (Tuscany, Italy), they highlight the insight that might be
gained from longitudinal studies at sites of greater clinical relevance such as the
proximal femur, distal radius and vertebral body.
2.1.2 Upper Extremity: Metacarpals, Radius and Ulna
Women have smaller bones in the upper extremity than men. Total area of the
metacarpals and radii are approx. 25 % less in women than men [12, 13], while
cortical area is approx. 35 % less in women than men [9].
Diaphyseal expansion with aging also occurs in long bones of the upper
extremity (Table 2). Analysis of hand radiographs revealed increased periosteal
and medullary diameters at the midshaft of the metacarpals [11, 12, 14]. Similar to
bones of the lower extremity, increases in periosteal diameter and sub-periosteal
area are modest (1–2 %/decade [11, 12]), while medullary diameter increases are
moderate (*15 %/decade in women, 5 %/decade in men [14]). The radius has
also been examined in several studies, using physical measurement [15], single
photon absorptiometry [16, 17] and, more recently, pQCT [9, 13, 18]. Most of the
cross-sectional studies reveal modest periosteal expansion with relatively greater
medullary expansion leading to a net loss of cortical bone area (*-3 %/decade)
in both women and men [9, 13, 17, 18]. In partial contrast, Burr et al. reported that
periosteal expansion and increased moment of inertia occurred in men but not in
women [15].
8
Table 2 Diaphyseal changes—upper extremity
Authors Materials Methods Study design Bone Sex D_P D_M TA MA CA Moment of inertia
a b
Garn et al. 2799 subjects PA CS, Metacarpal Female 0.7 0.9 – – – – –
[11] (1671 female, radiographs LNG a b
Male 0.5 0.7 – – – – –
1128 male)
25–84 years
c
Garn et al. 5660 subjects PA CS Metacarpal Female – – 1.7 – – –
[12] (3455 female, radiographs c
Male – – 0.6 – – –
2205 male)
25–85 years
Maggio 296 subjects PA CS Metacarpal Female 1.6 (NS) 15.5 – – – –
et al. [14] (189 female, radiographs Male 0.9 (NS) 5.3 – – – –
107 male)
30–88 years
Burr and 86 donors Bone sections CS Radius Female 0.3 (NS) – – – -6.8 -1.9 (NS)
Martin (male and (38 % site) Male 2.3 – – – 0.6 (NS) 6.0
[15] female)
18–95 years
d
Bouxsein 42 women CT (33 % site) CS Radius Female 1.0 (NS) – 0.8 (NS) 10.5 -2.6 0 (NS)
et al. [17] Young: Ulna Female 1.5 – 2.2 13.3 -2.4 3 (NS)
20-30 years
Old:
63–84 years
f c c c c
Ahlborg et al. 108 women Single photon LNG Radius Female 7 11 14 24 12 31
[16] *52– absorpt.
67 years e (6 cm)
(continued)
Table 2 continued
Authors Materials Methods Study design Bone Sex D_P D_M TA MA CA Moment of inertia
g
Kaji et al. 482 subjects pQCT (20 % CS Radius Female (NS) Increase – – Decrease –
[13] (252 female, site)
230 male) Male Increase Increase – – Decrease
Age-Related Changes
26–84 years
d
Yuen et al. 1258 subjects pQCT (33 % CS Radius Female – – – – -6.7 -3
[9] (638 female, site)
620 male) Male – – – – -2.5 -1.3
20–98 years
h i
Ward et al. 728 men pQCT (50 % CS Radius Male – – 0.8 (NS) 6.5 -2.7 –
[18] 40–79 years site)
D_P, periosteal diameter; D_M, medullary diameter; TA, total (subperiosteal) area; MA, medullary area; CA, cortical bone area
CS, cross-sectional study design; LNG, longitudinal study design
Metacarpal: 50 % site (mid-shaft); Radius/Ulna: section location is given as absolute distance or percent of limb length, relative to the distal end
CS, Cross-sectional study design; LNG, Longitudinal study design
a
based on cross-sectional study design (Table 1)
b
based on longitudinal analysis of subset of subjects (Table 2)
c
authors computed area assuming circular cross-section
d
e
baseline age is at individual menopause, with avg. 15 year follow-up
f
slopes based on individual regressions for each subject; value at menopause is reference for % change
g
age correlation: significance and sign reported but not slope
h
slopes were presented separately for two study centers (Table 3); avg. values shown here
i
cortical thickness is reported, not cortical area
9
Fig. 2 Longitudinal changes in total area (TA), marrow area (MA) and cortical bone area (CA)
of the tibial diaphysis as measured by pQCT at 3- and 6-year follow up (recreated from Fig. 1 of
Lauretani et al. [3], with permission). Subjects were grouped by decade at baseline scan. These
results indicate that temporal changes differed greatly with decade and between women and men,
not always in the manner indicated by the cross-sectional studies. For example, past age 80,
women had greater rates of increase in total area than men, both sexes had increasing marrow
area, and women appear to have stable bone area while men lost bone area
Greater rates of change at the distal radius were reported in a longitudinal study
of 108 women tracked from menopause until age 67 years (avg. post-menopausal
follow-up period 15 years) using single photon absorptiometry [16]. Areal mea-
sures were estimated from diameter measurements assuming a circular cross-
section. Subperiosteal area increased 14 %/decade, while medullary area increased
24 %. Because the absolute increase in subperiosteal area exceeded the absolute
increase in medullary area, there was a net increase in cortical bone of 12 %/
decade. There are several reasons why this study may have found greater rates of
change than other studies of the radius. It may reflect the study population of
Swedish women, as well as the focus on the 15-year interval after menopause,
which may be a period of relatively rapid bone turnover. It is also possible that the
longitudinal study design captured the true rates of change, which may have been
underestimated by cross-sectional studies. However, there are limitations to the
use of a two-dimensional projection method to estimate geometric changes in non-
circular bone cross-sections. Additional longitudinal studies using pQCT are
needed to clarify temporal changes in bone morphology in the upper extremity and
determine if there are indeed such high rates of periosteal and endosteal expansion
and net increases in cortical bone area.
2.2 Structural Mechanical Properties
Data on whole-bone mechanical properties for human long bone diaphyses are
remarkably few. There are some reports on femoral and tibial properties that were
used to provide reference values for the design of synthetic bones used in bio-
mechanical studies of fracture fixation [19]; however, donor age information is
incomplete and the focus is on elastic not failure properties. Martin and Atkinson
[4] estimated the bending strength of femoral shafts from 37 donors (22–82 years)
using beam theory along with direct measures of cross-sectional geometry and
indirect measures of material strength (based on bone density). Their results
suggested that female bones had a decline in strength in the second half of life,
attributed to modest declines in both material strength and section modulus (I/c),
while male bones maintained their strength with aging. The sample size in this
study was small, and it is not clear how accurate these estimates of bending
strength are. Notably, Russo et al. [6] reported similar trends for density-weighted
moment of inertia of the tibial diaphysis, i.e., that it declined slightly with age in
women (-2 %/decade) but not in men.
We identified only two studies that reported mechanical data on long bone
diaphyses across a range of ages, one for femur and one for radius. Martens et al.
[20] torsionally tested 46 femurs from donors aged 27–80 years (13 female, 33
male). Based on data reported in Table 1 of their paper, torsional rigidity (stiffness
normalized by specimen length) and failure moment did not correlate with age
(rigidity: p = 0.38, r2 = 0.02; moment: p = 0.095, r2 = 0.06), whereas fracture
energy (a measure of fracture resistance) was negatively correlated with age, albeit
weakly (p = 0.037; r2 = 0.10). This finding is consistent with the view that
diaphyseal stiffness does not change with age due to negligible changes in moment
of inertia and elastic modulus (a measure of stiffness at the tissue/material level),
whereas diaphyseal fracture energy declines with age due to decreases in bone
material toughness (see ‘‘Cortical Bone Mechanics and Composition: Effects of
Age and Gender’’).
Burr and Martin [15] torsionally tested 86 radii from donors (approx. equal
numbers female, male) aged 18–95 year. They reported only elastic properties.
Radii from females were approx. 40 % less stiff (torsional rigidity) than males.
With aging, the whole-bone stiffness did not change in females, whereas it
increased by approx. 6 %/decade in males. These changes were consistent with the
corresponding changes in moment of inertia (Table 2), whereas there were no
changes in estimated shear modulus.
Clearly, additional data at the whole-bone level are needed to determine how
changes in morphology and material properties influence diaphyseal fracture
resistance at different sites with aging.
2.3 Coordinated Variations in Geometric and Material

Properties in Mice and Men
Functional adaptation within the skeletal system can coordinate morphological

traits and tissue-level mechanical properties so the particular combination results
in a bone that is sufficiently stiff and strong to support the loads imposed during
daily activities [21]. The interaction among morphological traits and tissue-level
mechanical properties results in each person acquiring not a single trait, but a
particular set of traits. The particular set of traits acquired by a person will vary
with their height and body weight, as expected. Importantly, even when correcting
for body size there remains tremendous variation among individuals of the same
height and weight. Thus, bone size not only varies with body size [22, 23], but the
relationship between bone width and bone length (i.e., robustness) also varies
widely among individuals [24].
Because bone stiffness is proportional to the fourth power of diameter, the
natural variation in bone diameter [25] appears to be accompanied by coordinated
variations in cortical area and tissue-modulus in order to maximize the stiffness of
slender bones (narrow relative to length) and minimize the mass of robust bones
(wide relative to length). This phenomenon, which was consistent with a theo-
retical understanding of the functional adaptation process [26], was observed in
mouse long bone [27, 28] and translated to human long bone [29, 30] and also to
more complex, cortico-cancellous sites like the proximal femur [31, 32] and
vertebral body [33].
The variation in bone size relative to body size is likely to be compensated,

otherwise slender structures will be under-designed (weak) and robust structures
will be over-designed (bulky). For diaphyseal structures, slender bones show a
proportionally greater amount of cortical bone to maximize mass, and greater
mineralization to increase tissue-modulus [29, 30, 34, 35]. In contrast, robust
bones show proportionally less cortical area to minimize mass and lower miner-
alization. This functional adaptation process is not fully compensated, however.
Limitations in the degree to which bone cells are able to adjust tissue-modulus
results in slender tibiae being 2–3 times less stiff relative to body size compared to
robust tibiae [29]. The combination of naturally reduced stiffness plus the com-
pensatory increase in mineralization may help explain the increased risk of stress
fractures in military recruits and young-adult athletes having slender tibiae
[36–39]. Thus, the coordination between bone morphology and tissue-level
mechanical properties may be a critical factor responsible in establishing func-
tionality for habitual loading as well as fracture susceptibility.
Despite the fact that the periosteum is an important target for prophylactic
treatment, the amount of periosteal expansion required to maintain strength has
not been defined. Many factors contributing to variation in periosteal expansion
have been identified [14, 16, 40–46], including bone width [47]. In a theoretical
study, Lazenby showed that the amount of periosteal apposition depended on the
ratio of the inner to outer radii (i.e., r/R), which varies with external bone size
[29, 35]. In general, slender bones do not need to expand as much as robust bones
in order to maintain stiffness over time. The sex-specific differences in periosteal
expansion support this phenomenon, however, whether this size-dependence is
observed within a single sex has yet to be determined. Importantly, the age-related
decrease in tissue-modulus [48] would have to be compensated by increases in
periosteal expansion beyond that expected for endocortical resorption. The
magnitude of the interactions among age-related decreases in tissue-modulus,
the natural variation in bone diameter, and periosteal expansion have yet to be
established.
3 Metaphyseal and Vertebral Changes with Age
Metaphyseal sites differ from diaphyseal sites by the presence of trabecular bone in
the medullary cavity and by thinner cortices. They also differ in the much higher
incidence of osteoporotic/age-related fractures. Metaphyseal sites that are sus-
ceptible to osteoporotic fractures include the femoral neck and distal radius. The
vertebral bodies of the spine are analogous to metaphyseal sites given their
proximity to joints (i.e., interverbral discs) and their high proportion of trabecular
bone. The age-related changes at each of these fracture-prone sites has been
evaluated using in vivo imaging, primarily QCT. The distal tibia has also been
well studied, probably because of the ease of access for peripheral QCT more so
than clinical significance.
3.1 Cross-Sectional Geometry and Bone Mineral Density
Similar to the diaphyses, the metaphyses of long bones and the vertebral bodies
exhibit age-related periosteal and endosteal expansion. The average rates of
periosteal expansion are comparable at metaphyseal and diaphyseal sites (1–2 %/
decade). Because metaphyseal and vertebral sites have trabecular bone in the
medullary cavity, endosteal expansion corresponds to increased trabecular area.
The average rates of endosteal expansion are notably less at metaphyseal sites
(1–2 %/decade) than diaphyseal sites (5–15 %/decade).
The changes in morphology at metaphyseal and vertebral sites occur concur-
rently with dramatic reductions in bone density, as quantified by volumetric bone
mineral density (vBMD). vBMD values are often reported separately for trabecular
and total (trabecular plus cortical) regions. Cortical regions can be difficult to
reliably isolate because of the thin cortical shell and partial volume averaging
effects that are problematic for clinical QCT. Because we are presenting com-
parisons in terms of percent changes per decade rather than absolute changes, it is
difficult to compare the relative impact of cortical and trabecular bone loss.
Nonetheless, it is clear that both cortical and trabecular bone loss contribute
importantly to age-related bone loss at the femoral neck, distal tibia, distal radius,
and spine.
A longitudinal study by Riggs et al. [49] gives particular insight into the dif-
ferences in trabecular and cortical bone loss, and into the variable rates of change
during aging. They followed *1100 subjects (aged 20–97 years at baseline) for 3
years, obtaining QCT scans at the distal tibia, distal radius and lumbar spine.
Importantly, their results challenge some of the conventional wisdom about when
bone loss begins. Trabecular vBMD declines throughout life in women and men,
such that approximately 40 % of the lifetime loss of trabecular bone occurs before
age 50. By contrast, only 6 and 15 % of the lifetime loss of cortical bone occurs
before age 50 in women and men, respectively. Notable decreases in cortical
vBMD do not begin until around the time of menopause in women and even later
in life in men.
3.1.1 Lower Extremity: Femoral Neck and Distal Tibia
The femoral neck is not accessible to pQCT, but because of the clinical sig-
nificance of hip fractures, age effects at the femoral neck have been examined
using DXA and clinical QCT (Table 3). The size (total area) of the femoral neck
is 25–30 % smaller in women than men [8, 50]. In both sexes, average increases
in total area are 1.5 %/decade, with increases in medullary area of 2.5 %/decade.
As at other sites, these changes result in cortical bone loss (i.e., decreased
cortical area) that is somewhat greater in women (-5 %/decade) than men
(-3 %/decade). Trabecular bone loss at this site is dramatic, and is greater in
women than men; average declines in trabecular vBMD based on two reports are
Table 3 Metaphyseal changes—lower extremity
Authors Materials Method Study design Bone Sex TA MA CA Tot.vBMD Trab.vBMD
a
Russo et al. [6] 1205 subjects pQCT CS Distal Tibia Female 0.8 – – -4.9 -4.6
(693 female, Male 1.8 – – -0.7 -2.9
512 male)
Age-Related Changes
20–102 years
b
Riggs et al. [50] 696 subjects pQCT CS Distal Tibia Female 0.3 (NS) 0.9 -2.3 -5.3 -4.0
(373 female, Male 0.6 (NS) 0.9 (NS) -0.3 -3.6 -4.5
323 male)
20–97 years
c
Lauretani et al. [3] 809 subjects pQCT CS, LNG Distal Tibia Female – – – -5.0 -3.7
(464 female, Male – – – -2.9 -3.3
345 male)
21–102 years
a, d
Yuen et al. [9] 1258 subjects pQCT CS Distal Tibia Female 0.7 1.8 – – -6.4
(638 female, Male 0.8 2.7 – – -6.0
620 male)
20–98 years
c e
Macdonald et al. [51] 644 subjects HR-pQCT CS Distal Tibia Female 1.1 – -3.5 -6.5 -2.8
(442 female, e
Male 0.9 – -1.3 -3.4 -2.3
202 male)
20–99 years
a f g h i
Duan et al. [67] 1196 subjects DXA CS Femoral Neck Female 0.5 1.5 -5.6 -6.2 –
(801 female, f g h i
Male 1.9 2.7 -4.2 -5.0 –
395 male)
18–92 years
b
Riggs et al. [50] 696 subjects QCT CS Femoral Neck Female 1.8 3.8 -2.3 -8.9 -11.7
(373 female, Male 1.0 2.2 -4.0 -6.0 -8.5
323 male)
20–97 years
15
(continued)
Table 3 (continued)
16

Sigurdsson et al. [8] 1715 subjects QCT CS Femoral Neck Female 1.8 – -6.3 -8.5 -22.6
(908 female, Male 2.1 – -1.7 -2.6 -14.2
807 male)
67–93 years
TA, total (subperiosteal) area; MA, medullary area (called trabecular area in some papers); CA, cortical bone area; Tot.vBMD, avg. bone mineral density of
total volume; Trab.vBMD, average bone mineral density of trabecular volume
All results based on cross-sectional (CS) study design
Distal Tibia: 4 % site; location given as percent of limb length, relative to the distal end; Femoral Neck: mid-portion of neck
a
b
c
d
e
Trab.BV/TV reported, not Trab.vBMD
f
Periosteal diameter is reported, not TA
g
Endocortical perimeter is reported, not MA
h
Cortical thickness is reported, not CA
i
BMC/estimated volume
-17 %/decade in women and -11 %/decade in men [8, 50]. The rate of decline
in total vBMD is intermediate to the rates of decline of cortical and trabecular
bone, and is significantly greater in women (-9 %/decade) than men (-5 %/
decade in men).
The distal tibia is accessible to peripheral QCT scanning, and several recent
studies report modest age-related periosteal and endosteal expansion, with cortical
thinning and net loss of cortical area in both women and men (Table 3). Overall,
the distal tibia in women is smaller and has lower BMD than in men, differences
that are present at young adulthood and persist with aging [3, 6, 9, 50, 51]. In both
sexes, the increase in total area is only about 1 %/decade, with similar increases in
medullary area. The average cortical bone loss is slightly greater in women
(-3 %/decade) than men (-1 %/decade). Trabecular bone density declines in
women and men by -3 %/decade. The combined effect of cortical and trabecular
loss results in declines in total vBMD averaging -5 %/decade in women and
-2.5 %/decade in men, much slower than at the proximal femur.
The longitudinal study by Riggs et al. [49] gives insight into the rates of change
of bone density at the distal tibia, early and later in life. Somewhat surprisingly, the
rate of trabecular bone loss is identical in pre- and post-menopausal women
(-2.4 %/decade). In men, the rate of trabecular bone loss is actually greater before
age 50 (-4 %/decade) than after age 50 (-1.7 %/decade). Cortical bone exhibits a
different pattern. There is no decline in cortical vBMD in pre-menopausal women,
but a -3.6 %/decade decline in post-menopausal women. In men the rates of
change are not significantly different before and after age 50 (approx.
-1 %/decade), although the rate increases sharply after age 80 (-3.9 %/decade).
3.1.2 Upper Extremity: Distal Radius
The distal radius is of clinical relevance as a common site of osteoporotic frac-

tures, and is accessible by peripheral QCT. The age-related changes at this site are
similar to changes at the distal tibia, albeit a bit more pronounced, especially at
later ages (Table 4). In brief, average bone size (i.e., total area) at the distal radius
is 25–30 % smaller in women than men [9, 50–52]. Most studies indicate modest
periosteal and endosteal expansion in both sexes, with net cortical bone loss
marginally greater in women (-2 %/decade) than men (-1 %/decade). Several
studies indicate loss of trabecular and total vBMD that is comparable in women
and men (approx. -5 %/decade) [9, 50, 51]. One study that examined an older age
range (60–90 years) reported much greater rates of decline in cortical thickness
(-11 %/decade women; -4 %/decade men) and trabecular BV/TV (-11 %/
decade women; -5 %/decade men) [52].
Similar to the distal tibia, the rate of trabecular bone loss at the distal radius is
not different between pre- and post-menopausal women (approx. -5 %/decade), or
between men before and after age 50 (-4 %/decade). The rate of cortical bone loss
is negligible in pre-menopausal women and younger men, but is significant in
older women (-5 %/decade) and men (-3 %/decade) [49].
Table 4 Metaphyseal changes—upper extremity
18

a
Riggs et al. [50] 696 subjects (373 female, pQCT CS Distal Radius Female 0.9 2.3 -1.6 -5.1 -4.0
323 male) 20–97 years Male 2.1 3.2 -0.2 -5.0 -4.5
Mueller et al. [52] 100 specimens HR-pQCT CS Distal Radius Female NS – -10.9 b -11.6c -11.4 c
(50 female, 50 male) 60–100 years Male NS – -4.1 b -5.6 c -6.2 c
d, e
Yuen et al. [9] 1258 subjects pQCT CS Distal Radius Female 2.0 3.8 – – -8.2
(638 female, 620 male) 20–98 years Male 2.0 3.6 – – -7.0
f
Macdonald et al. [51] 644 subjects (442 female, 202 male) HR-pQCT CS Distal Radius Female 0.2 – -3.0 -6.0 -3.6
20–99 years Male 2.7 – -1.3 -5.2 -4.2
g
Ward et al. [18] 728 men 40–79 years pQCT CS Distal Radius Male 1.2 – – -4.5 -3.1
TA, total (subperiosteal) area; MA, medullary area (called trabecular area in some papers); CA, cortical bone area; Tot.vBMD, avg. bone mineral density of total
volume; Trab.vBMD, average bone mineral density of trabecular volume
Distal Radius 4 % site; location given as percent of limb length, relative to the distal end
a
b
Cortical thickness reported
c
BV/TV reported
d
e
f
g
slopes were presented separately for two study centers (Table 3); avg. value shown here
3.1.3 Spine
Similar to other bones, vertebrae of women are smaller than men. Based on QCT
data, the cross-sectional area of the vertebral body is 20–25 % less in women than
men [8, 50, 53].
The phenomenon of age-related periosteal expansion is also observed in the
vertebral bodies of the spine in both sexes (Table 5). Ruhli et al. measured linear
dimensions in a sample of spines from 71 ‘‘modern’’ donors (30 female, 41 male)
and reported that vertebral diameter increased modestly with age in men (e.g.,
sagittal diameter of C7 increased by approx. 3 %/decade) but not in women [54].
Mosekilde and Mosekilde reported a 3 % increase per decade in the area of the L2
vertebral body from a pooled sample of cadavers from males and females [55].
Using QCT, Riggs et al. reported a similar modest increase in area of approx. 2 %/
decade in both sexes [50]. In another study using QCT, Sigurdsson et al. reported
periosteal expansion of approx. 5 %/decade in both sexes, notably greater than
values from other studies or other sites.
Loss of trabecular bone is pronounced in the vertebral body, with average
declines of -18 %/decade in women and -13 %/decade in men (Table 5). The
decline in total vBMD is also significantly greater in women (-12 %/decade)
than men (-6 %/decade). Riggs et al. [49] reported that trabecular bone loss in
the lumbar spine accelerates with age, which is different than their findings at the
distal tibia and radius (reviewed above). Trabecular vBMD declined at a rate of
-16 %/decade in pre-menopausal women and -26 %/decade in post-meno-
pausal women. The rate of loss was slower in men, but also increased with age
(-8.4 %/decade before age 50; -18.5 %/decade after age 50).
3.2 Metaphyseal and Vertebral Bone Strength
There are surprisingly few studies of metaphyseal and vertebral strength at the
whole-bone level that provide data on age-related changes in mechanical prop-
erties. Mechanical testing of cadaveric specimens is the gold standard for this
information, but such studies are challenging because it is difficult to collect
enough samples across a large age range from young to old (e.g., 20-90 years).
Moreover, many investigators have focused on evaluating how well metrics from
DXA and QCT imaging (e.g., BMD) correlate to bone strength, and may have
included age as a covariate but not reported age-dependence per se. We review the
available data on whole-bone mechanical properties and age in this section.
In the past decade, CT-based finite element analysis (FEA) methods have been
shown to be accurate enough to allow surrogate assessment of bone strength using
virtual analysis. In the ‘‘continuum’’ approach developed by several groups (e.g.,
Keyak et al., Keaveny et al.), a QCT image set is used to generate a computer finite
element model that incorporates geometric and densitometric information and can
be used to perform a virtual mechanical test. If properly validated by comparison
Table 5 Vertebral changes—lumbar spine
20

Authors Materials Method Study Bone Sex TA Tot.vBMD Trab.vBMD
design
Mosekilde & Mosekilde 44 donors (27 female, 17 male) Physical CS L2 Vertebra Pooled 2.9 – –
[55] 15–87 years measurement
a
Ebbesen et al. [53] 101 donors (51 female, 50 male) QCT CS L3 Vertebra Female – – -19
18–96 years Male – – -15
c
Duan et al. [68] 719 donors b (477 female, 242 DXA CS L3 Vertebra Female 0.9 -11.6 –
male) Male 2.3 -4.3 –
18–92 years
d
Riggs et al. [50] 696 subjects (373 female, 323 QCT CS L1–L3 Female 1.9 -12.2 -10.9
male) Vertebra Male 1.9 -7.6 -9.7
20–97 years
Sigurdsson et al. [8] 1715 subjects (908 female, 807 QCT CS L1–L2 Female 4.7 -13.3 -25.1
male) Vertebra Male 6.0 -6.3 -14.9
67–93 years
TA, total (subperiosteal) area; Tot.vBMD, avg. bone mineral density of total volume; Trab.vBMD, average bone mineral density of trabecular volume
a
estimated from Fig. 3 of paper
b
includes young and elderly groups, not fracture groups
c
authors estimated CSA and vBMD values from two DXA projections; change based on two-point slope from data in their Table 1
d
of model predictions to mechanical properties in cadaveric samples, these methods

open the door to analyze samples from many hundreds of subjects from popula-
tion-based and clinical trials. A variant on this approach is to use microCT to
generate a voxel-based model which accurately renders the micro- as well as the
macro-structure. These microCT-based FEA models have been used most widely
to generate models of small (*1 cm) trabecular samples, but in recent years have
been used to generate whole-bone models, most often using HR-pQCT images of
the distal radius. Either of these FEA approaches produces an estimate of bone
strength. We review the available data on age and estimated whole-bone
mechanical properties from CT-based FEA in this section.
3.2.1 Lower Extremity: Femoral Neck and Distal Tibia
Loading the proximal end of the femur in a configuration replicating a fall to the
side is the loading mode most relevant to hip fractures, which are fractures of the
proximal femur and nearly always occur as a result of a fall to the side [56, 57].
There are limited experimental data on age and mechanical properties of the
proximal femur/femoral neck. Courtney et al. [58] compared femora from young
donors (17–51 yrs, mean 33 yrs; n = 9) to femora from old donors (59–83 yrs,
mean 74 yrs; n = 8) using fall-configuration loading. The old femurs had 30 %
lower stiffness, 50 % lower maximum force (‘‘strength’’) and required 70 % less
energy to fracture compared to the young femurs. This corresponds to a decrease in
strength of -12 %/decade. Limitations of this study include the grouping of female
(n = 7) and male (n = 10) donors, and the small sample size which did not allow
for linear regression analysis. In another experimental study using a falls config-
uration, Roberts et al. [59] tested 73 elderly cadaveric femora (range 55–98 yrs,
mean 74 yrs; 48 female, 25 male). They did not focus on the effects of age, but did
report a weak, non-significant correlation between failure load and age (r = -0.14;
p = 0.08; sexes pooled); separate regression analyses for female and male were not
reported. This result suggests that there is only a weak association of age and
proximal femur failure force from middle- to old-age. In a recent study, Epelboym
et al. [31] subjected 49 female cadaveric proximal femora (29–93 years of age) to a
failure load in a simulated fall-to-the side, and reported that maximum load
(r2 = 0.19, p \ 0.004) but not stiffness (r2 = 0.05, p \ 0.18) decreased with age.
While the load-age regression was only moderately strong, the effects are more
dramatic when comparing subsets of young versus old subjects. A comparison of
properties between women less than 50 years of age (n = 6) to women older than
80 years of age (n = 13) revealed a 35 % reduction in stiffness, a 46 % reduction in
maximum load, and a 50 % reduction in work-to-fracture. A multiple regression
analysis indicated that cortical area and trabecular BMD were the most significant
contributors to the variation in maximum load.
One limitation in accurately determining age-related changes from mechanical
test data is that they are based on relatively small samples. By contrast, studies that
use QCT-based FEA have access to many more samples. Keaveny et al. [60]
generated QCT-based FEA models of the proximal femur from 362 women and
317 men (aged 21–89 years) from the Rochester, MN study group of Riggs et al.
[50], and reported that estimated femoral strength (fall configuration) was strongly
age dependent. Femoral strength declined by 55 % in women and 39 % in men.
The age-dependence was roughly bi-modal, with negligible changes until the mid-
40s for women and the mid-50s for men, followed by sharp declines (Fig. 3a).
Notably, the declines in strength accelerated with increasing age and were much
greater than the corresponding declines in areal BMD (Fig. 3b). The authors noted
that the deficit in strength in older women compared to older men was due mostly
to a delay in the onset of decline for men rather than differences in initial values or
rates of decline. We note that the strong, linear decline in strength after middle-age
is at odds with the experimental findings from Roberts et al. [59], who found a
non-significant decline after middle-age. This discrepancy may be due to differ-
ences in techniques (experimental vs. FEA) or to differences in samples (cadavera
from donors of unknown racial/ethnic makeup vs. a random population sample
[98 % white] from Rochester, MN). Additional experimental work is needed to
clarify this discrepancy.
In a recent longitudinal study, Lang et al. [61] used QCT-based FEA to estimate
changes in failure force of the proximal femur over a 5-year period in 112 elderly
women and 111 elderly men (avg. age at baseline 77 years). Loading conditions
were applied to stimulate stance and fall configurations, using methods previously
validated by Keyak and co-workers [62]. At baseline, the failure load in a fall
configuration was 29 % lower for women than men, and in a stance configuration
was 25 % less for women than men. These sex differences are essentially the same
as the sex differences in proximal femur size described in the previous section.
Women had significantly greater declines in trabecular BMD and strength than
men. In the fall configuration, the rate of change in strength was -25 %/decade in
women compared to -14 %/decade in men. These declines can be attributed to
loss of both trabecular and cortical bone, although the rates of trabecular loss were
much greater.
The difference in average bone strength between women and men has been
noted in all the studies we reviewed, and is consistently attributed to size differ-
ences. But do women and men ‘‘build’’ bones of similar structure? A recent study
by Srinivasan et al. [63] used QCT-based FEA models to test whether women and
men who have similar femoral neck areal BMD values have similar femoral
strengths. They reported that in 114 women and 114 men with matched aBMD, the
equivalence of aBMD was the result, on average, of greater bone size in men
(38 % greater bone area) combined with lower volumetric bone density (16 %
lower vBMD). Estimated femoral bone strength in 28 women and 28 men matched
for aBMD showed similar values of bone strength and load-to-strength ratio in the
two groups. Thus, based on this relatively small sample, it appears that the
proximal femur of women and men have different structural ‘‘designs’’ that can
accomplish similar function (strength).
Compared to the steep declines in estimated strength of the proximal femur,
declines in strength of the distal tibia are much less. MacDonald et al. [51]
Fig. 3 Change in estimated femoral strength with age (from Keaveny et al. [60], with
permission). Strength was predicted from finite element models simulating impact from a
sideways fall on the proximal femur. a Predicted strength declines after age 50 in women and
after age 60 in men. When presented as mean values for each decade, the rates of decline are
linear and appear similar in women and men. b The annualized change in predicted strength and
areal BMD reveal much greater rates of decline in strength than BMD, and also reveal
accelerating percentage declines in strength with increasing age
estimated strength of the distal tibia using HR-pQCT-based FEA in 442 women
and 202 men (20–99 years). In women, strength declined approx. -6 %/decade
versus -3 %/decade in men. Thus, although accessibility to pQCT imaging makes
it an attractive site for study, the distal tibia does not appear to have the dramatic
age-related declines in strength that occur at the proximal femur.
3.2.2 Upper Extremity: Distal Radius
Several studies have reported age-related decreases in bone strength of the distal
radius based either on mechanical testing or FEA. Bonel et al. [64] performed
compression tests on forelimbs from 25 female and 24 male cadavers (57–100 years).
The majority of specimens (73 %) failed by fracture of the distal radius. In this group,
bones from females were 30 % weaker than from males, which corresponded to a
25 % smaller bone size (TA). Both sexes had evidence of age-related decline in bone
strength, although the change in females was significant (-14.4 %/decade;
p \ 0.05) whereas the change in males was not (-6.6 %/decade; p [ 0.05).
Mueller et al. [52] examined the distal radius from 50 female (mean age 82 yrs)
and 50 male (mean age 80 yrs) cadavers using HR-pQCT and mechanical testing that
produced Colles-type fractures. They scanned the distal 20 % of the bone, but noted
that failure force correlated most strongly with trabecular morphology in the most
distal 4 % (‘‘region 1’’ in their paper); results reviewed here are from that region.
Bones from women were smaller, less dense and less strong than men. Total tissue
volume was 30 % less, cortical bone volume 42 % less, total BV/TV 25 % less, and
trabecular BV/TV 32 % less in women compared to men. Failure force was 38 % less
in women than men. With aging from 60–100 years, there was no evidence of
periosteal expansion, although both sexes lost bone. Both total and trabecular BV/TV
declined by -11 %/decade in women and -6 %/decade in men (Table 4). Failure
force declined -13 %/decade in women and -9 %/decade in men. Failure force was
strongly correlated with trabecular BV/TV (r2 = 0.74) Of note, the absolute rates of
decline were not significantly different in women and men, although the percent-per-
decade magnitudes tend to be greater in women, most likely because they have lower
absolute baseline values. In summary, the distal radii of women are smaller, less
dense and weaker than men; both sexes lose bone density and bone strength with
aging. Bone strength in an axial loading setup that produces a clinically relevant
fracture mode (Colles-type) correlates most strongly with trabecular bone volume
fraction at the distal-most region, indicating that trabecular bone loss is most relevant
to diminished fracture resistance of the distal radius.
MacDonald et al. [51] used HR-pQCT to generate FE models of the distal
radius in 425 women and 199 men (20–99 years). The median estimated failure
load was 40 % less in women than men, nearly identical to the difference mea-
sured by Mueller et al. [52]. The strength difference corresponded to 30 % smaller
bone size in women along with 25 % lower trabecular BV/TV. With aging, esti-
mated failure load decreased approx. -8%/decade in women and -5 %/decade in
men. These rates are less than those reported by Mueller et al., which may reflect
the wider age range in the MacDonald study group.
3.2.3 Spine
Mosekilde and Mosekilde [55] performed compression tests on thoracic and

lumbar vertebral bodies from 44 cadavers (27 female, 17 male; 15–87 years, mean
58 years). Whole-bone failure force was normalized by cross-sectional area, and
this average failure stress demonstrated a highly significant age-related decrease of
-15 %/decade (relative to value at 58 years; female and male pooled). We note
that the rate of decrease in whole-bone failure force would likely be slightly less
than this because of the 3 %/decade increase in cross-sectional area also reported
in this study.
Ebbesen et al. [53] performed compression tests on L3 vertebral bodies from
101 cadavers (51 female, 50 male; 18–96 yrs, mean 57 yrs). Vertebral whole-bone
strength in females was 20 % less than in males, which corresponded to the 20 %
gender difference in cross-sectional area. Both sexes exhibited age-related declines
in whole-bone strength of approx. -16 %/decade, and the rates of decline
(absolute slope of force vs. age regression) did not differ between the sexes.
Notably, the rates of decrease in force are nearly identical to the rates of decline of
trabecular bone density (Table 5). When expressed as failure stress (force divided
by cross-sectional area), there were no gender differences in average values or in
rates of decline with age. Thus, the authors concluded that differences in vertebral
strength between women and men are the result of differences in size, and that
women and men lose vertebral strength at the same rate with aging.
Bouxsein et al. [65] estimated vertebral compressive strength from QCT
parameters of *700 subjects from a Rochester, Minn. study group (21–97 yrs).
The age-related changes in area and density from this population were described
by Riggs et al. [50] and are summarized in Table 5. Using empirical relations
relating density and cross-sectional area to whole-bone vertebral strength, esti-
mated strength was observed to be significantly less in women than men, con-
sistent with differences in bone size. In slight contrast to the study of Ebbesen
et al., the rate of decline in strength was greater in women than men. Failure force
was 20 % less in young women than young men, and this difference increased to
30 % by age *85 years. Women lost strength at a rate of approx. -7 %/decade
compared to -4.5 %/decade for men. The difference in rates was attributed to the
significantly greater rate of decline of vBMD in women.
A recent study by Christiansen et al. [66] used QCT-based FEA to estimate
strength of T10 and L3 vertebral bodies in 120 subjects, equally divided between
young (35–42 yrs) women and men, and old (73–83 yrs) women and men. The
findings are consistent with those cited above in two regards: 1) young women and
men have differences in bone strength (approx. 15 % lower in women) that are
explained by differences in bone size rather than density; 2) age-related loss of
bone strength is entirely attributed to loss of bone density. However, the findings
differ notably from Ebbesen et al. in that the rate of trabecular bone loss with aging
was significantly greater in women than men, and consequently the rate of decline
in bone strength was significantly greater in women than men. Women lost bone
strength at an estimated -16 %/decade whereas men lost strength at –6.5 %/
decade (L3 and T10 averaged; values based on estimated mean strength).
In summary, there is consensus that at young ages vertebral strength is 15–20 %
lower in women than men, as a result of differences in bone size not bone density.
With aging, changes in strength parallel the loss of vertebral bone density; there is
little to no compensatory increase in vertebral cross-sectional area. The available
literature differ on the relative rates of decline in bone density between women and
men and thus in the rates of decline in bone strength. One study reported similar rates
of decline [53], whereas two others reported greater rates of decline in women than
men [60]. The lack of agreement likely reflects differences in donors/study popu-
lation, and highlights the variability that exists in patterns of age-related bone loss.
4 Conclusions
There is a wealth of data on changes in bone structure and BMD with age. Results
often differ between skeletal sites and between sexes. In addition, differences
between sample populations likely contribute to the variability in results between
studies. Nonetheless, there are some patterns that are seen in most studies.
Diaphyseal periosteal expansion occurs at all sites studied and is typically
observed in both women and men. Medullary expansion also occurs, typically at a
faster rate in women than men, resulting in net loss of cortical bone area in women,
but negligible loss in men. At metaphyseal sites there is also periosteal expansion
with endosteal expansion and net loss of cortical bone. But the most dramatic age-
related change at the metaphysis is loss of trabecular bone, which is observed in
women and men often at rates exceeding 10 %/decade. In some studies the rate of
trabecular bone loss is similar between the sexes, while in others women lose bone
at a faster rate than men. Data on the corresponding changes in whole-bone
mechanical properties with age are surprisingly few, especially for the diaphysis.
Increasingly, FEA is being used as surrogate for mechanical testing of the prox-
imal femur, distal radius and vertebra. Available data (mechanical and FEA)
indicate declines in bone strength with age that often correlate strongly with loss of
trabecular bone, especially at the distal radius and vertebra. Additional mechanical
data from large samples of bones covering 20–90 years would contribute greatly to
our understanding of the magnitude of strength loss with age, and how much of
this loss is explained by loss of cortical versus trabecular bone.
References
1. Burge, R., Dawson-Hughes, B., Solomon, D.H., Wong, J.B., King, A., Tosteson, A.: Incidence
and economic burden of osteoporosis-related fractures in the United States, 2005–2025.
J. Bone Miner. Res. 22, 465–475 (2007)
2. Feik, S.A., Thomas, C.D., Clement, J.G.: Age-related changes in cortical porosity of the
midshaft of the human femur. J. Anat. 191(Pt 3), 407–416 (1997)
3. Lauretani, F., Bandinelli, S., Griswold, M.E., Maggio, M., Semba, R., Guralnik, J.M.,
Ferrucci, L.: Longitudinal changes in BMD and bone geometry in a population-based study.
J. Bone Miner. Res. 23, 400–408 (2008)
4. Martin, R.B., Atkinson, P.J.: Age and sex-related changes in the structure and strength of the
human femoral shaft. J. Biomech. 10, 223–231 (1977)
5. Ruff, C.B., Hayes, W.C.: Sex differences in age-related remodeling of the femur and tibia.
J. Orthop. Res. 6, 886–896 (1988)
6. Russo, C.R., Lauretani, F., Bandinelli, S., Bartali, B., Di Iorio, A., Volpato, S., Guralnik, J.M.,
Harris, T., Ferrucci, L.: Aging bone in men and women: beyond changes in bone mineral
density. Osteoporos. Int. 14, 531–538 (2003)
7. Russo, C.R., Lauretani, F., Seeman, E., Bartali, B., Bandinelli, S., Di Iorio, A., Guralnik, J., Ferrucci,
L.: Structural adaptations to bone loss in aging men and women. Bone 38, 112–118 (2006)
8. Sigurdsson, G., Aspelund, T., Chang, M., Jonsdottir, B., Sigurdsson, S., Eiriksdottir, G.,
Gudmundsson, A., Harris, T.B., Gudnason, V., Lang, T.F.: Increasing sex difference in bone
strength in old age: the age, gene/environment susceptibility-Reykjavik study (AGES-
REYKJAVIK). Bone 39, 644–651 (2006)
9. Yuen, K.W., Kwok, T.C., Qin, L., Leung, J.C., Chan, D.C., Kwok, A.W., Woo, J., Leung, P.C.:
Characteristics of age-related changes in bone compared between male and female reference
Chinese populations in Hong Kong: a pQCT study. J. Bone Miner. Metab. 28, 672–681 (2010)
10. Smith Jr, R.W., Walker, R.R.: Femoral expansion in aging women: implications for
osteoporosis and fractures. Science 145, 156–157 (1964)
11. Garn, S.M., Rohmann, C.G., Wagner, B., Ascoli, W.: Continuing bone growth throughout
life: a general phenomenon. Am. J. Phys. Anthropol. 26, 313–317 (1967)
12. Garn, S.M., Frisancho, A.R., Sandusky, S.T., McCann, M.B.: Confirmation of the sex
difference in continuing subperiosteal apposition. Am. J. Phys. Anthropol. 36, 377–380 (1972)
13. Kaji, H., Kosaka, R., Yamauchi, M., Kuno, K., Chihara, K., Sugimoto, T.: Effects of age, grip
strength and smoking on forearm volumetric bone mineral density and bone geometry by
peripheral quantitative computed tomography: comparisons between female and male.
Endocr. J. 52, 659–666 (2005)
14. Maggio, D., Pacifici, R., Cherubini, A., Simonelli, G., Luchetti, M., Aisa, M.C., Cucinotta, D.,
Adami, S., Senin, U.: Age-related cortical bone loss at the metacarpal. Calcif. Tissue Int. 60,
94–97 (1997)
15. Burr, D.B., Martin, R.B.: The effects of composition, structure and age on the torsional
properties of the human radius. J. Biomech. 16, 603–608 (1983)
16. Ahlborg, H.G., Johnell, O., Turner, C.H., Rannevik, G., Karlsson, M.K.: Bone loss and bone
size after menopause. N. Engl. J. Med. 349, 327–334 (2003)
17. Bouxsein, M.L., Myburgh, K.H., van der Meulen, M.C., Lindenberger, E., Marcus, R.: Age-
related differences in cross-sectional geometry of the forearm bones in healthy women.
Calcif. Tissue Int. 54, 113–118 (1994)
18. Ward, K.A., Pye, S.R., Adams, J.E., Boonen, S., Vanderschueren, D., Borghs, H., Gaytant, J.,
Gielen, E., Bartfai, G., Casanueva, F.F., Finn, J.D., Forti, G., Giwercman, A., Han, T.S.,
Huhtaniemi, I.T., Kula, K., Labrie, F., Lean, M.E., Pendleton, N., Punab, M., Silman, A.J.,
Wu, F.C., O’Neill, T.W.: Influence of age and sex steroids on bone density and geometry in
middle-aged and elderly European men. Osteoporos. Int. 22, 1513–1523 (2011)
19. Heiner, A.D., Brown, T.D.: Structural properties of a new design of composite replicate
femurs and tibias. J. Biomech. 34, 773–781 (2001)
20. Martens, M., van Audekercke, R., de Meester, P., Mulier, J.C.: The mechanical
characteristics of the long bones of the lower extremity in torsional loading. J. Biomech.
13, 667–676 (1980)
21. Currey, J.D.: Bones: structure and mechanics. Princeton University Press, Princeton (2002)
22. Moro, M., van der Meulen, M.C., Kiratli, B.J., Marcus, R., Bachrach, L.K., Carter, D.R.:
Body mass is the primary determinant of midfemoral bone acquisition during adolescent
growth. Bone 19, 519–526 (1996)
23. Sumner, D.R., Andriacchi, T.P.: Adaptation to differential loading: comparison of growth-
related changes in cross-sectional properties of the human femur and humerus. Bone 19,
121–126 (1996)
24. Pearson, O.M.: Activity, climate, and postcranial robusticity: implications for modern human
origins and scenarios of adaptive change. Curr Anthropol 41, 569–607 (2000)
25. Pandey, N., Bhola, S., Goldstone, A., Chen, F., Chrzanowski, J., Terranova, C.J., Ghillani, R., Jepsen,
K.J.: Interindividual variation in functionally adapted trait sets is established during postnatal growth
and predictable based on bone robustness. J. Bone Miner. Res. 24, 1969–1980 (2009)
26. Currey, J.D., Alexander, R.M.: The thickness of the walls of tubular bones. J. Zool. (London)
206, 453–468 (1985)
27. Jepsen, K.J., Akkus, O.J., Majeska, R.J., Nadeau, J.H.: Hierarchical relationship between
bone traits and mechanical properties in inbred mice. Mamm. Genome 14, 97–104 (2003)
28. Jepsen, K.J., Hu, B., Tommasini, S.M., Courtland, H.W., Price, C., Terranova, C.J.,
Nadeau, J.H.: Genetic randomization reveals functional relationships among morphologic
and tissue-quality traits that contribute to bone strength and fragility. Mamm. Genome 18,
492–507 (2007)
29. Jepsen, K.J., Centi, A., Duarte, G.F., Galloway, K., Goldman, H., Hampson, N., Lappe, J.M.,
Cullen, D.M., Greeves, J., Izard, R., Nindl, B.C., Kraemer, W.J., Negus, C.H., Evans, R.K.:
Biological constraints that limit compensation of a common skeletal trait variant lead to
inequivalence of tibial function among healthy young adults. J. Bone Miner. Res. 26,
2872–2885 (2011)
30. Tommasini, S.M., Nasser, P., Schaffler, M.B., Jepsen, K.J.: Relationship between bone
morphology and bone quality in male tibias: implications for stress fracture risk. J. Bone
Miner. Res. 20, 1372–1380 (2005)
31. Epelboym, Y., Gendron, R.N., Mayer, J., Fusco, J., Nasser, P., Gross, G., Ghillani, R., Jepsen, K.J.:
The inter-individual variation in femoral neck width is associated with the acquisition of
predictable sets of morphological and tissue-quality traits and differential bone loss patterns.
J Bone Miner Res (2012). doi:10.1002/jbmr.1618 [Published online: 28 Mar, 2012]
32. Zebaze, R.M., Jones, A., Knackstedt, M., Maalouf, G., Seeman, E.: Construction of the
femoral neck during growth determines its strength in old age. J. Bone Miner. Res. 22,
1055–1061 (2007)
33. Tommasini, S.M., Hu, B., Nadeau, J.H., Jepsen, K.J.: Phenotypic integration among
trabecular and cortical bone traits establishes mechanical functionality of inbred mouse
vertebrae. J. Bone Miner. Res. 24, 606–620 (2009)
34. Tommasini, S.M., Nasser, P., Jepsen, K.J.: Sexual dimorphism affects tibia size and shape but
not tissue-level mechanical properties. Bone 40, 498–505 (2007)
35. Tommasini, S.M., Nasser, P., Hu, B., Jepsen, K.J.: Biological co-adaptation of morphological
and composition traits contributes to mechanical functionality and skeletal fragility. J. Bone
Miner. Res. 23, 236–246 (2008)
36. Beck, T.J., Ruff, C.B., Mourtada, F.A., Shaffer, R.A., Maxwell-Williams, K., Kao, G.L.,
Sartoris, D.J., Brodine, S.: Dual-energy x-ray absorptiometry derived structural geometry for
stress fracture prediction in U.S. Marine Corps recruits. J. Bone Min. Res. 11, 645–653 (1996)
37. Crossley, K., Bennell, K.L., Wrigley, T., Oakes, B.W.: Ground reaction forces, bone
characteristics, and tibial stress fracture in male runners. Med. Sci. Sports Exerc. 31,
1088–1093 (1999)
38. Giladi, M., Milgrom, C., Simkin, A., Stein, M., Kashtan, H., Margulies, J., Rand, N., Chisin, R.,
Steinberg, R., Aharonson, Z., et al.: Stress fractures and tibial bone width. A risk factor. J. Bone
Joint Surg. Br. 69, 326–329 (1987)
39. Milgrom, C., Giladi, M., Simkin, A., Rand, N., Kedem, R., Kashtan, H., Stein, M., Gomori,
M.: The area moment of inertia of the tibia: a risk factor for stress fractures. J. Biomech.
22, 1243–1248 (1989)
40. Lanyon, L., Armstrong, V., Ong, D., Zaman, G., Price, J.: Is estrogen receptor alpha key to
controlling bones’ resistance to fracture? J. Endocrinol. 182, 183–191 (2004)
41. Lee, K., Jessop, H., Suswillo, R., Zaman, G., Lanyon, L.: Endocrinology: bone adaptation
requires oestrogen receptor-alpha. Nature 424, 389 (2003)
42. Midura, R.J., Su, X., Morcuende, J.A., Tammi, M., Tammi, R.: Parathyroid hormone rapidly
stimulates hyaluronan synthesis by periosteal osteoblasts in the tibial diaphysis of the
growing rat. J. Biol. Chem. 278, 51462–51468 (2003)
43. Rivadeneira, F., Houwing-Duistermaat, J.J., Vaessen, N., Vergeer-Drop, J.M., Hofman, A.,
Pols, H.A., Van Duijn, C.M., Uitterlinden, A.G.: Association between an insulin-like growth
factor I gene promoter polymorphism and bone mineral density in the elderly: the Rotterdam
study. J. Clin. Endocrinol. Metab. 88, 3878–3884 (2003)
44. Turner, R.T., Backup, P., Sherman, P.J., Hill, E., Evans, G.L., Spelsberg, T.C.: Mechanism of
action of estrogen on intramembranous bone formation: regulation of osteoblast
differentiation and activity. Endocrinology 131, 883–889 (1992)
45. Uusi-Rasi, K., Sievanen, H., Vuori, I., Pasanen, M., Heinonen, A., Oja, P.: Associations of
physical activity and calcium intake with bone mass and size in healthy women at different
ages. J. Bone Miner. Res. 13, 133–142 (1998)
46. Zhai, G., Rivadeneira, F., Houwing-Duistermaat, J.J., Meulenbelt, I., Bijkerk, C., Hofman, A.,
van Meurs, J.B., Uitterlinden, A.G., Pols, H.A., Slagboom, P.E., van Duijn, C.M.: Insulin-like
growth factor I gene promoter polymorphism, collagen type II alpha1 (COL2A1) gene, and the
prevalence of radiographic osteoarthritis: the Rotterdam study. Ann. Rheum. Dis. 63, 544–548
(2004)
47. Lazenby, R.A.: Continuing periosteal apposition. II: The significance of peak bone mass,
strain equilibrium, and age-related activity differentials for mechanical compensation in
human tubular bones. Am. J. Phys. Anthropol. 82, 473–484 (1990)
48. Burstein, A.H., Reilly, D.T., Martens, M.: Aging of bone tissue: mechanical properties.
J Bone Jt Surg Am 58, 82–86 (1976)
49. Riggs, B.L., Melton, L.J., Robb, R.A., Camp, J.J., Atkinson, E.J., McDaniel, L., Amin, S.,
Rouleau, P.A., Khosla, S.: A population-based assessment of rates of bone loss at multiple
skeletal sites: evidence for substantial trabecular bone loss in young adult women and men.
J. Bone Miner. Res. 23, 205–214 (2008)
50. Riggs, B.L., Melton, L.J., Robb, R.A., Camp, J.J., Atkinson, E.J., Peterson, J.M., Rouleau, P.A.,
McCollough, C.H., Bouxsein, M.L., Khosla, S.: Population-based study of age and sex
differences in bone volumetric density, size, geometry, and structure at different skeletal sites.
J. Bone Miner. Res. 19, 1945–1954 (2004)
51. Macdonald, H.M., Nishiyama, K.K., Kang, J., Hanley, D.A., Boyd, S.K.: Age-related patterns
of trabecular and cortical bone loss differ between sexes and skeletal sites: a population-based
HR-pQCT study. J. Bone Miner. Res. 26, 50–62 (2011)
52. Mueller, T.L., van Lenthe, G.H., Stauber, M., Gratzke, C., Eckstein, F., Muller, R.: Regional,
age and gender differences in architectural measures of bone quality and their correlation
to bone mechanical competence in the human radius of an elderly population. Bone 45,
882–891 (2009)
53. Ebbesen, E.N., Thomsen, J.S., Beck-Nielsen, H., Nepper-Rasmussen, H.J., Mosekilde, L.:
Age- and gender-related differences in vertebral bone mass, density, and strength. J. Bone
Miner. Res. 14, 1394–1403 (1999)
54. Ruhli, F.J., Muntener, M., Henneberg, M.: Age-dependent changes of the normal human
spine during adulthood. Am J Hum Biol 17, 460–469 (2005)
55. Mosekilde, L., Mosekilde, L.: Normal vertebral body size and compressive strength: relations
to age and to vertebral and iliac trabecular bone compressive strength. Bone 7, 207–212 (1986)
56. Hayes, W.C., Myers, E.R., Morris, J.N., Gerhart, T.N., Yett, H.S., Lipsitz, L.A.: Impact near
the hip dominates fracture risk in elderly nursing home residents who fall. Calcif. Tissue Int.
52, 192–198 (1993)
57. Grisso, J.A., Capezuti, E., Schwartz, A.: Falls as risk factors for fractures. In: Marcus, R., Feldman, D.,
Kelsey, J. (eds.) Osteoporosis edition), pp. 599–611. Academic Press, San Diego (1996)
58. Courtney, A.C., Wachtel, E.F., Myers, E.R., Hayes, W.C.: Age-related reductions in the strength
of the femur tested in a fall-loading configuration. J Bone Jt Surg Am 77, 387–395 (1995)
59. Roberts, B.J., Thrall, E., Muller, J.A., Bouxsein, M.L.: Comparison of hip fracture risk prediction
by femoral aBMD to experimentally measured factor of risk. Bone 46, 742–746 (2010)
60. Keaveny, T.M., Kopperdahl, D.L., Melton 3rd, L.J., Hoffmann, P.F., Amin, S., Riggs, B.L.,
Khosla, S.: Age-dependence of femoral strength in white women and men. J. Bone Miner.
Res. 25, 994–1001 (2010)
61. Lang, T.F., Sigurdsson, S., Karlsdottir, G., Oskarsdottir, D., Sigmarsdottir, A., Chengshi, J.,
Kornak, J., Harris, T.B., Sigurdsson, G., Jonsson, B.Y., Siggeirsdottir, K., Eiriksdottir, G.,
Gudnason, V., Keyak, J.H.: Age-related loss of proximal femoral strength in elderly men and
women: The Age Gene/Environment Susceptibility Study—Reykjavik. Bone 50, 743–748 (2012)
62. Keyak, J.H., Rossi, S.A., Jones, K.A., Skinner, H.B.: Prediction of femoral fracture load using
automated finite element modeling. J. Biomech. 31, 125–133 (1998)
63. Srinivasan, B., Kopperdahl, D.L., Amin, S., Atkinson, E.J., Camp, J., Robb, R.A., Riggs, B.L.,
Orwoll, E.S., Melton 3rd, L.J., Keaveny, T.M., Khosla, S.: Relationship of femoral neck areal
bone mineral density to volumetric bone mineral density, bone size, and femoral strength in
men and women. Osteoporos. Int. 23, 155–162 (2012)
64. Bonel, H.M., Lochmuller, E.M., Well, H., Kuhn, V., Hudelmaier, M., Reiser, M., Eckstein, F.:
Multislice computed tomography of the distal radius metaphysis: relationship of cortical bone
structure with gender, age, osteoporotic status, and mechanical competence. J Clin Densitom
7, 169–182 (2004)
65. Bouxsein, M.L., Melton 3rd, L.J., Riggs, B.L., Muller, J., Atkinson, E.J., Oberg, A.L.,
Robb, R.A., Camp, J.J., Rouleau, P.A., McCollough, C.H., Khosla, S.: Age- and sex-
specific differences in the factor of risk for vertebral fracture: a population-based study
using QCT. J. Bone Miner. Res. 21, 1475–1482 (2006)
66. Christiansen, B.A., Kopperdahl, D.L., Kiel, D.P., Keaveny, T.M., Bouxsein, M.L.:
Mechanical contributions of the cortical and trabecular compartments contribute to
differences in age-related changes in vertebral body strength in men and women assessed by
QCT-based finite element analysis. J. Bone Miner. Res. 26, 974–983 (2011)
67. Duan, Y., Beck, T.J., Wang, X.F., Seeman, E.: Structural and biomechanical basis of sexual
dimorphism in femoral neck fragility has its origins in growth and aging. J. Bone Miner. Res.
18, 1766–1774 (2003)
68. Duan, Y., Turner, C.H., Kim, B.T., Seeman, E.: Sexual dimorphism in vertebral fragility is
more the result of gender differences in age-related bone gain than bone loss. J. Bone Miner.
Res. 16, 2267–2275 (2001)
Characterisation of Trabecular Bone
Structure
Ian H. Parkinson and Nicola L. Fazzalari
Abstract The characterisation of trabecular bone structure has until recently

relied on morphometric analysis of histological sections although there is now
wide availability of bench top non-destructive X-ray-based imaging with the
ability to resolve trabecular elements at resolution on the order of *10 microns.
The advent of non-destructive X-ray-based imaging, such as micro-computed
tomography (micro-CT) has enabled measurements from image datasets, repre-
senting the three-dimensional structure of trabecular bone. Ex vivo studies into
trabecular bone structure in osteoporosis have mainly focused on clinically rele-
vant skeletal sites, such as the proximal femur, the distal radius and vertebral
bodies. In vivo, the iliac crest and the sternum have been used to obtain material
for the diagnosis of metabolic bone diseases including osteoporosis. Metaphyseal
bone structure is determined early in development as secondary trabeculae emerge
from the primary spongiosa in the epiphyseal plates during endochondral bone
growth. After closure of the epiphyseal growth plates at skeletal maturity, bone
remodelling becomes the predominant means by which bone is added or removed
from the trabecular compartment. From the time of attainment of peak bone mass,
studies show that there is a decrease in trabecular bone volume through to older
age in both sexes, although not at all sites and not uniformly for males and
females. Gender specific changes in trabecular bone are most evident at and after
the menopause in females, which is associated with decreased estrogen and
associated with reduced androgen production in males. The consequence of
I. H. Parkinson (&) N. L. Fazzalari

Bone and Joint Research Laboratory, SA Pathology,
Adelaide, SA, Australia
e-mail: ian.parkinson@health.sa.gov.au
I. H. Parkinson N. L. Fazzalari
Discipline of Anatomy and Pathology, University of Adelaide,
Adelaide, SA, Australia

DOI: 10.1007/8415_2011_113
Published Online: 12 January 2012
32 I. H. Parkinson and N. L. Fazzalari
menopausal or age-related bone loss for females and males, respectively, is a

marked increase in fracture incidence, although the changes to the trabecular bone
architecture are different between sexes.
1 Introduction
Trabecular bone is found at the end of the long bones of the appendicular skeleton
and in the vertebral bodies of the axial skeleton. The bone has a complex, porous
spatial arrangement and the spatial complexity contributes to maximal strength for
minimum mass for the skeleton as a whole [23]. The high mineral surface area
associated with the arrangement of the trabecular bone elements provides a vast
substrate on which cellular interaction with bone mineral material can occur.
The characterisation of trabecular bone structure has until recently relied on
morphometric analysis of histological sections. While histological sections only
provide a two-dimensional ‘‘snapshot’’ of the complex three-dimensional entity,
the insights gained from these preparations have been validated and further
developed by the data analysis from three-dimensional imaging modalities that
have become the methods of choice for studying trabecular bone [12, 27, 75, 78].
The main difference between these methodologies is that of bias in estimating the
dimensions of the trabecular bone structure from histological sections [78]. The
bias originates from the use of Parfitt’s idealized plate and rod model of trabecular
bone structure [85]. The destructive nature of histological section preparation has
limited the study of bone strength to parallel investigations in cross-sectional
studies [2, 19, 20, 24, 28, 34, 37, 70, 78, 86, 89, 98, 104, 106].
There is now wide availability of bench top non-destructive X-ray-based imaging
with the ability to resolve trabecular elements at resolution on the order of
*10 microns. The advent of non-destructive X-ray-based imaging, such as micro-
computed tomography (micro-CT) has enabled measurements from image datasets,
representing the three-dimensional structure of trabecular bone [63, 79, 100, 101].
Subsequent mechanical testing of the same sample has enabled explanatory models
of bone strength to be developed, which provide further understanding of the change
in trabecular bone structure associated with ageing and disease [6, 64, 102]. These
three-dimensional datasets have also been used as input for finite element analysis
models to determine apparent mechanical properties of bone [62, 81]. To the present,
trabecular bone has been studied at multiple skeletal sites, in all age groups from
neonates to the elderly [4, 14, 42, 111]. Advantages given by non-destructive
imaging of trabecular bone include the ability to subsequently perform mechanical
testing on the bone samples, histological analysis and/or gene expression analysis
[6, 105, 106].
Ex vivo studies into trabecular bone structure in osteoporosis have mainly
focused on clinically relevant skeletal sites, such as the proximal femur, the distal
radius and vertebral bodies [3, 35, 36]. In vivo, the iliac crest and the sternum have
been used to obtain material for the diagnosis of metabolic bone diseases including
Characterisation of Trabecular Bone Structure 33
osteoporosis [69, 70, 89, 91], but ethical issues and the availability of clinical non-
invasive imaging have greatly reduced this source of material. Societal sensitivi-
ties and government regulation have also restricted the availability of cadaveric
material, which has tended to constrain these studies to an older age range thus
limiting the clinical relevance of studies [34, 36].
Metaphyseal bone structure is determined early in development as secondary
trabeculae emerge from the primary spongiosa in the epiphyseal plates during
endochondral bone growth [14, 42]. Modelling of individual bones in childhood
and adolescence occurs through periosteal apposition and endosteal resorption to
change the size and shape of the cortical shell of the bone and the trabeculae in the
trabecular bone compartment; these appeat to adapt in a coordinated manner to
maintain the ability to withstand the extant loads [45, 82, 83, 109].
After closure of the epiphyseal growth plates at skeletal maturity, bone
remodelling becomes the predominant means by which bone is added or removed
from the trabecular compartment [45]. It is now well established that the archi-
tecture of trabecular bone is dependent on the forces acting upon it and the high
surface area of the bone mineral in the bone tissue marrow facilitates the cellular
events, which remove or deposit bone in a highly dynamic environment [45].
Coordination of the cellular events may be at the tissue level for events such as
removal of damaged bone [82, 83], at the organ level for modelling due to changed
usage pattern, or the whole body level for involutional events such as the meno-
pause in females [90]. The consequences of these scenarios at all size scales can be
to decrease the load needed to cause a fracture, where in-built safety margins in
load carrying capacity are reduced [8].
From the time of attainment of peak bone mass, studies show that there is a
decrease in trabecular bone volume with aging in both sexes [68], although not at all
sites and not uniformly for males and females [9]. The specific cause of these age-
related changes in individuals is unclear as most studies were cross-sectional in
nature making it difficult to track the many factors that influence bone mass, such
as the loading history of the individuals. While it is known that trabecular bone
architecture changes according to the loading history of the individual, there are
other factors, such as nutrition, co-morbidities, social activities and work activities
that affect bone metabolism, independent of direct mechanical stimuli [54].
In females, changes in trabecular bone are most evident at and after the men-
opause, which is associated with decreased estrogen [22]. The greatly increased
activation of osteoclasts associated with decreased oestrogen in menopausal
females results in an imbalance between resorption and formation with a conse-
quent net bone loss [90]. In males, the reduction in sex-hormone (androgen)
production is typically more gradual but is associated with significant net bone loss
over time as a consequence of increased resorption relative to formation [55, 103].
The consequence of menopausal or age-related bone loss for females and males,
respectively, is a marked increase in fracture incidence, although the changes to
the trabecular bone architecture are different between sexes [43, 109]. In general,
menopausal females lose trabecular bone through perforation of trabeculae, which
are then either completely removed or transformed from plate-like to rod-like
Fig. 1 X-ray image of a

sagittal slice from the
proximal femur, which shows
the complex patterns formed
by trabeculae
[1, 70]. In males, age-related bone loss occurs through gradual thinning of tra-
beculae, where the connectivity of the structure is largely intact but overall the
structure has a reduced capacity to withstand load [1, 22, 101]. Similar findings
have also been reported in micro-CT studies, which show that the bias in the
estimation of the magnitude of trabecular dimensions using histological sections
does not mask relative differences between groups [99].
2 Skeletal Distribution of Trabecular Bone: What and Where

to Sample
Trabecular bone is found at the end of the medullary cavities of hollow long bones
throughout the skeleton (Fig. 1). It forms a trabecular network, of interconnected
rod-like and plate-like structures, which are found in greater or lesser proportion
depending on the skeletal site [4, 85]. Interestingly, the skeletal sites with the
greatest volume of trabecular bone are, in general, the sites where the majority of
osteoporotic or fragility fractures occur [52, 53]. Hence, there has been and con-
tinues to be a strong focus on understanding the contribution of trabecular bone to
mechanical strength of the bone as an organ and in how therapeutic intervention
prevents loss of mechanical integrity [6, 10, 71, 94].
Fig. 2 X-ray image of sagittal femoral head slice (left) in contrast to a macerated sagittal slice
from a vertebral body (right) showing marked differences in the amount of bone and the
arrangement of trabeculae
Histological studies have shown, as expected, that there is considerable heter-

ogeneity in the trabecular microstructure at clinically relevant sites [4, 46]. For
example, bone volume fraction varies from less than 7% in the centrum of ver-
tebral bodies to more than 25% in regions in the femoral head [3, 35] (Fig. 2).
Consequent to this variability in bone volume fraction, the trabecular architecture
is highly variable, not just in terms of how plate-like or rod-like it is but in the
dimensions of the trabecular elements and how they are arranged in space [46–48].
Eckstein evaluated trabecular microstructure at seven sites (distal radius, L2
vertebra, femoral neck, femoral trochanter, iliac crest, calcaneus) in 79 female and
86 male cadavera, and reported that correlation coefficients between sites were
weak to moderate (0.01–0.56). For example, BV/TV at one site explained only
10–32% of the variance of BV/TV at other sites.
The amount of trabecular bone adjacent to joints through which large forces are
transmitted suggests that it plays an important role in maintaining the mechanical
integrity of the joint complex and in the mechanical dynamics of the body a whole.
Optimization of the micro-architecture of the trabecular network in response to extant
loads imparts optimal strength with minimal mass [23, 41], with inbuilt safety margins
[8]. The large mineral surface area afforded by the complex trabecular network pro-
vides a vast substrate on which directed cellular activity can interact with the bone
mineral material [13]. This enables bone to be removed or laid down very rapidly in a
coordinated manner in response to biomechanical or physiological signals [93].
As stated previously, there is considerable heterogeneity in the distribution of
trabecular bone within skeletal sites [4, 98] and between skeletal sites [4].
Therefore, where a sample is taken is extremely important as to its relevance to the
experimental questions that can be answered. In relation to studies of osteoporotic
bone it is necessary that the samples have a structure that is equivalent to the
structure where a fracture may have occurred or is at increased risk of occurring.
Fig. 3 Composite photomicrograph showing histological sections of trabecular bone from

multiple skeletal sites
Availability of human trabecular bone samples has always and continues to be a

problematical issue in experimental design. For ex vivo cadaveric studies, samples
have been obtained from clinically-relevant skeletal sites, such as the proximal
femur, the distal radius and vertebral bodies. These samples have enabled the
morphology of the trabecular bone at sites where osteoporotic fractures occur to be
characterised and for comparative studies (Fig. 3) [49–51]. Change in bone mor-
phology over time, between sexes or between clinically-relevant groups can be
determined [34, 92]. For in vivo studies, biopsies from the iliac crest and the sternum
have been obtained because of ease of accessibility to these sites without the need for
general anesthesia. These biopsies have enabled changes in bone morphology or
cellular dynamics to be investigated from subsequent histomorphometric analysis.
It has become increasingly difficult to obtain human bone material due in part to
changing societal attitudes, human ethics considerations for in vivo studies and
clinical imperative. For cadaveric ex vivo studies, regulatory frameworks, in the
western world, are at best highly restrictive but where access to human cadaveric
material is allowed informed consent from next of kin is at the centre of all
requests for material [5]. These restrictions have had the consequence of generally
limiting the age range of study cohorts to the older end of the human age-range,
which while still clinically relevant it does not encompass all clinical groups.
Surgically-sourced bone samples for ex vivo studies can be obtained through direct
patient consenting protocols but rely upon motivated surgical, nursing and medical
liaison staff. For in vivo studies, and where humans ethics approval for human
experimentation is permitted, biopsies are most useful for monitoring changes in
cellular dynamics and bone material properties in response to therapeutic inter-
ventions. In longitudinal studies, serial biopsies are obtained over an appropriate
timescale [10]. In the past, iliac crest biopsies in particular, were useful in the
diagnosis and characterisation of metabolic disorders affecting bone, including
osteoporosis [21, 70]. However, advances in interpretation of bone serology and
imaging have reduced the clinical imperative for in vivo bone biopsy [43, 103].
3 Histomorphometry of Trabecular Bone Structure
From the 1960s until the mid 1990s, histological sections were the most widely
used tool for the study of trabecular bone structure. Pioneers in this field, such as
Harold Frost and Michael Parfitt, developed histomorphometric techniques to
enable the characterisation of trabecular bone micro-architecture as well as the
cellular dynamics of bone [39, 85].
Protocols have been developed for resin embedding of undecalcified bone
samples to enable thin sections (less than 10 micron thick when cut in the
longitudinal direction) to be cut thus preserving the bone mineral phase. Such
undecalcified sectioning allows bone components to be visualized by utilizing the
physical chemistry of bone mineral, where a modified von Kossa technique [15]
localizes the mineralized bone phase and haematoxylin and eosin (H&E) is used to
stain the unmineralized or osteoid bone phase (Fig. 4). The H&E stain also enables
visualization of bone cells, such as osteoclasts and osteoblasts at the bone surfaces
in the same sections. Surfaces at which active mineralisation is occurring are
localized by sequential administration of fluorescing compounds, which are
incorporated, in vivo, at sites of mineralizing bone [39]. Together these techniques
have provided a platform on which to conduct cross-sectional biopsy-based clin-
ical studies or ex vivo laboratory-based studies.
In addition to observational descriptions of bone, quantitative protocols were
developed through the adaptation of stereological techniques [40, 56, 86] specif-
ically for the complex architecture of trabecular bone. Bone histomorphometry
uses a suite of structural descriptors, which were developed based on idealized
models of trabecular bone structure as plates, rods or mixed plates and rods [85].
Independent bone tissue measures such as bone mineral area, bone tissue area
and bone mineral perimeter are applied to stylized models of trabecular bone
structure and indices describing the average architectural properties are derived.
These indices include independent measures of bone volume per tissue volume
Fig. 4 Histological images of osteoclasts removing bone matrix (left) and osteoblasts laying
down osteoid (right). Von Kossa’s silver impregnation with Hematoxylin and Eosin counterstain.
(Black is bone mineral, on the left panel, arrows indicate actively resorbing osteoclasts and on the
right panel arrows indicate osteoblast actively laying down osteoid on bone surfaces)
(BV/TV, also called ‘bone volume fraction’), bone surface per tissue volume
(BS/TV) and bone surface per bone volume (BS/BV). From these independent
measures, derived parameters include trabecular thickness (Tb.Th), trabecular sep-
aration (Tb.Sp) and trabecular number (Tb.N) [87]. These histomorphometric
indices of bone structure have been universally adopted [84] and continue to be used.
Bone histomorphometry has also been extended to bone cell dynamics,
including mineralization measured from fluorochrome labeling. By measuring the
extent of bone surface with osteoid, resorption pits or active mineralization it is
possible to derive indices of bone turnover [39, 106]. These techniques have
provided insights into the temporal sequence of the components of the basic
multicellular unit (BMU) [39], such that activation frequency for new BMUs can
be calculated, the extent of resorption measured, the extent of osteoid measured
and the rate of mineralization of the osteoid calculated [13]. More recently, the
principles of bone histomorphometry have been adapted to quantify microdamage
accumulation, where en bloc or bulk staining of whole bone samples preferentially
stains areas of microdamage, which can be visualized after subsequent processing
into resin and sectioning [30, 32, 72, 95].
The quantitative techniques available for study of trabecular bone structure and
trabecular bone cellular dynamics encompass manual, semi-quantitative/interac-
tive and automated techniques. Manual quantitation from histological sections of
trabecular bone involves point counting, where a grid pattern overlaid on the
section determines where sampling of the feature of interest occurs [56]. These
point counts provide estimates of area or perimeter of the features of interest and
by application of Parfitt’s plate or rod model to the raw point counts, indices
representing the trabecular bone structure are derived [84]. Point counting tech-
niques have been applied to provide static indices of trabecular bone structure,
such as bone volume fraction, trabecular thickness, trabecular separation and
trabecular number as well as dynamic indices of bone structure, such as bone
formation rate and bone apposition rate [40]. Semi-quantitative techniques typi-
cally utilize a camera-mounted microscope interfaced to a computer monitor
where customised or commercial (Bioquant, Bioquant Image Analysis Corpora-

tion, Nashville, TN; Osteomeasure, Osteometrics Inc., Decatur, GA) software
enables features of interest to be delineated in spatial registration with the
microscope image. Static and dynamic indices characterising the trabecular
structure are calculated based on the same stereological principles employed for
point counting techniques [18]. Automated quantitation of trabecular bone from
histological sections has mainly been confined to a small range of static parameters
derived from bone area, bone perimeter and tissue area measurements [86].
Essentially, only bone matrix can be reliably segmented by automated techniques
in histological sections, hence only global static indices of bone structure are
obtained; however, when performed in conjunction with manual or semi-quanti-
tative techniques all descriptive bone parameters can be obtained.
The importance of these techniques developed over the past 50 years is
underlined by their continued use today. While currently available imaging
technology and techniques have extended the ability to visualize and quantify
trabecular bone structure they have not replaced manual and interactive techniques
but have become complementary to them.
4 Non-Destructive Imaging and Morphometry

of Trabecular Bone
Notwithstanding the conceptual models developed to extrapolate measurements

from histological sections to represent the three-dimensional micro-architecture of
trabecular bone, the field has always strived to develop and adapt technology to
enable true 3D analysis of this complex entity. The availability of benchtop high-
resolution micro-computed tomography (micro-CT) in the mid 1990s was enthu-
siastically welcomed by the bone community [37, 46]. For ex vivo studies micro-
CT scanners provide isotropic spatial resolution on the order of 10 microns albeit
for samples less than 10 mm in diameter [76, 77]. However, for bone samples up
to 50 mm in diameter 15 micron spatial resolution is achievable, which enables
the thinnest trabecular elements (70 microns in diameter) to be resolved (Fig. 5)
[77]. The non-destructive nature of micro-CT imaging means that complementary
methodologies can be applied to the samples, subsequent to imaging. For example,
conventional histology can be performed if cellular dynamics are of interest [78],
or genetic analysis can be performed to identify genes associated with bone dis-
eases [31, 58] or mechanical testing can be performed to enable predictive models
of bone strength to be formulated [7, 38, 44, 62, 88, 102].
While laboratory-based micro-CT has become ubiquitous for quantitative bone
studies there are other non-destructive imaging modalities available to researchers.
Synchrotron facilities provide X-ray tomography but with a monochromatic X-ray
source, which provides much better delineation between the mineral and non-
mineral phases in bone than the polychromatic X-ray source used in laboratory-
based micro-CT imaging [11, 17]. However, access to synchrotron facilities, while
Fig. 5 3D rendering of a bone cube from the L4 vertebral body of 66 years old male with
BV/TV = 9.6%
improving, is problematic and there are limited facilities in which large clinically-
relevant bone samples can be imaged. In addition, synchrotron-derived datasets
present challenges in data handling, where it is not unusual for a single sample to
generate at least 100 GB of data, whereas a dataset from a laboratory-based micro-
CT system will be less than 10 GB in size. These massive synchrotron datasets
require computing resources not commonly available. For non-ionizing radiation
imaging, magnetic resonance (MR) imaging is approaching the spatial resolution
of micro-CT, where in-plane resolution is approaching 100 lm and apparent
resolution below 100 lm with sub-voxel processing techniques [59]. However,
compared to micro-CT imaging there are still limitations in the ability to accu-
rately delineate the mineral and non-mineral phases, which have limited the
adoption of this imaging modality for morphometric bone studies [57, 59, 61, 67].
One of the challenges of three-dimensional imaging of trabecular bone by
whatever imaging modality is the ability to manipulate the wide range and large
volume of data acquired from imaging, for 3D reconstruction and for morpho-
metric analysis. However, consumer-level computers are able to process desktop
micro-CT datasets reasonably efficiently, in terms of processing time, data gen-
eration and data storage needs.
Fig. 6 Colour-coded 3D-rendered image of trabecular bone from the intertrochanteric region of
the femur showing decomposition of the trabecular structure into individual trabeculae elements.
(colour coding of individual trabeculae provides visual contrast between trabeculae and is not
indicative of morphology)
The availability of 3D voxel-based datasets of trabecular bone has driven the

development of quantitative tools to validate insights gained from histological
studies and to extend morphometric capabilities to more realistic representations
of the 3D structure of trabecular bone. Through implementation of an algorithm
that fit spheres to 3D datasets, ‘‘real’’ measures of trabecular diameter and tra-
becular separation are possible [47, 48]. Together with algorithms that describe
how plate-like or rod-like the structure is (Structure model index, SMI) [47, 48],
whether there is preferential orientation of the structure (Degree of anisotropy,
DA) [112] or how well connected the structure is (Connectivity Density. Conn.
D) [80]. Suites of histomorphometric measures are available within commercially
available micro-CT systems. More recently, in parallel, Stauber et al. [100, 101]
and Liu et al. [63] have developed algorithms that volumetrically ‘‘decompose’’
the trabecular structure into individual elements, which are then classified as rods
or plates (Fig. 6). These tools provide the size, shape and orientation of the
individual trabeculae, which enables study into how individual trabecular mor-
phology or orientation contributes to the mechanical properties of the structure as
a whole [65, 102].
Fig. 7 Scanning electron microscope image of endochondral bone in human neonate rib (left),
and graph showing change in cartilage volume fraction and bone volume fraction (y-axis) versus
distance from the resting zone to the mid-metaphysis (x-axis) (right) [33] with permission
5 Trabecular Bone Structure in the Young
Although there have been relatively few studies describing trabecular bone
structure in development and adolescence, the morphological events in bone
growth are well described [14, 33, 42] (Fig. 7) and the rate of acquisition of
bone mass has been shown to take place relatively slowly until puberty when the
rate significantly increases [45]. The velocity of bone growth is different between
boys and girls, reflecting differences in onset of puberty and in response to dif-
ferences in musculature and while there is convergence in growth rates towards
adulthood, clear sex-related size differences in the skeleton are maintained
throughout life [45, 109].
There is less information as to the relative importance of the genetic template
for bone structure versus the effects of environmental factors, i.e., nature versus
nurture. In the neonate growth plate histological studies clearly show that the
spatial arrangement of the columnar hypertrophic chondrocytes give rise to pri-
mary spongiosa and hence the secondary trabeculae [14, 42]. Whether the quality
of adult trabecular bone structure is determined or influenced at this early stage has
been hypothesized [33], although the ability of bone to adapt to the prevailing
mechanical environment shows that genetic influence cannot fully determine an
adult’s bone microarchitecture.
There is a peak in fracture incidence in the young around the time of puberty,
which in girls is approximately 11.5–12.5 years, and in boys is approximately
13.5–14.5 years [45]. These fractures are commonly associated with moderate
trauma, with the majority occurring in the distal radius. While trabecular bone
structure has not been directly implicated as contributing to susceptibility to
fracture it has been shown that during adolescence there is a transient increase in
longitudinal and circumferential growth of the cortex before there is a corre-
sponding increase in bone mass through thickening of the cortex and consolidation
of the trabecular bone structure within the medullary space [45, 60].
Notwithstanding the genetic or environmental effects on trabecular bone

structure prior to adulthood [45], future susceptibility to osteoporotic fractures will
be minimised, for an individual, if consideration is given to maximizing bone
health during these formative years. It has been shown that the skeleton is capable
of very high calcium absorption during growth [45], which allows for rapid
modeling and optimization of the trabecular architecture, particularly at sites
where the majority of osteoporotic fractures occur.
6 Trabecular Bone Structure in the Adult
Peak bone mass and thus, maximal bone strength is attained late in the 3rd decade
of life [45]. Complete cessation of endochondral bone formation means that further
change in trabecular bone structure is primarily through remodeling of existing
trabeculae. Regulation of the bone basic multicellular unit (BMU) through the
coordinated action of osteoclasts and osteoblasts has been well characterised at
morphological, physiological and molecular levels, which has provided insights
into how trabecular bone is maintained, repaired and remodeled in response to
physiological or mechanical stimuli. Histomorphometric studies have shown that
the sequence of cellular events that occur when bone is removed and subsequently
replaced by unmineralized matrix (osteoid) can be quantified in time and space and
in vivo fluorochrome labeling has enabled the rate of mineralization of the osteoid
to be measured. Together these measured parameters provide a detailed snapshot of
an individual’s bone metabolism at the site of sampling. It has been suggested that
remodeling can be targeted or untargeted [82, 83], where targeted remodelling is
most usually a reparative process in response to damage accumulation [13, 82, 83],
which is initiated by disturbance to the cannilicular network as microcracks pro-
gress within the bone matrix. The amount of bone turnover has been shown to be in
excess of the that required to maintain mechanical competence therefore it is has
been suggested that, while initially targeted, there is some remodeling that con-
tinues even when the initiating stimulus is no longer present [82, 83].
From the 4th to 6th decades studies have shown that trabecular bone volume
fraction can decline by up to 40–50% for males and females [1, 45, 68, 73]
although the rate of decline is sex-dependent, site-dependent and study cohort
dependent. An exception to this general pattern is in lactation, where up to 10% of
bone mass is lost in response to the nutritional imperative of milk production,
where the bone loss is mediated by mammary gland-derived parathyroid hormone
related-protein (PTHrP) in combination with low estrogen levels [16]. Fortunately,
this insult to the skeleton is transient with rapid restoration of bone mass after
weaning and it is not thought to infer greater susceptibility to fracture later in life.
In clinically relevant skeletal sites, there are significant differences in the
bone volume fraction of trabecular bone between males and females [4], which
are associated with differences in the trabecular microstructure [1]. There is
also considerable variability in trabecular microstructure within skeletal sites
Fig. 8 3D rendering from micro-CT images of trabecular bone samples from human vertebral
bodies of two individuals, which have similar bone volume fraction but differ markedly in
architecture. (both samples were obtained from individuals over the age of 60, where the BV/TV
was 9.3 and 8.6%, respectively)
[1, 29, 98, 110, 113], particularly at sites with large load-bearing. During the 4th
to 6th decades and particularly before the menopause in women, the prevalence
of low-impact or fragility fractures is low compared to older age groups.
However, the incidence of fractures at sites such as the distal radius, the ribs and
ankles rises significantly after the age of 35 [108].
7 Trabecular Bone Structure in Older Age
Large and clinically relevant changes in trabecular bone structure occur from the
6th decade of life onward (Fig. 8; Table 1). The Rotterdam study [96] shows that
incidence of non-vertebral fractures in osteopenic and osteoporotic males and
females (diagnosed based on BMD t-scores) more than doubles after the 7th
decade. Fracture risk is site dependent and males and females have different dis-
tributions of prevalence in sites of fracture, for example the incidence rate per
1,000 person years for hip fractures in males is 3.0, whereas the incidence rate per
1,000 person years for females is 6.9 [96]. The age at which particular skeletal
sites show increased incidence of fractures differs between the sexes, for example
the incidence of distal radius fractures in females increases markedly from the age
of 55, whereas in men the incidence of these fracture does not increase until after
the age of 75 [96].
In females, there is accelerated loss of bone mass from the onset of menopause,
which can be within the 5th decade but more usually in the 6th decade. Cessation
of estrogen production removes an important control on osteoclast activation,
Table 1 Changes in trabecular bone structure or density with ageing from middle to old age
Study Materials Site Male versus Percent change per decade
(parameter) female
Female Male
QCT [25] 51 F, 50 M L3 Vertebra ns -9% -9%
19–96 years (vBMD)
Micro-CT 75 F, 75 M L2 Vertebra ns -9.1 ns
[26, 66] 52–99 years (BV/TV)
Iliac crest ns -15.5 ns
(BV/TV)
Femoral neck F\M -13.4 ns
(BV/TV) (-35%)
Femoral troch. F\M ns 9.0
(BV/TV) (-19%)
Calcaneus ns -8.3 ns
(BV/TV)
Distal radius F\M ns ns
(BV/TV) (-30%)
HR-pQCT 64 F, 66 M Distal radius F\M -11.4 -9.1
[74] 60–99 years [region 1] (-32%)
(BV/TV)
F female M male; ns not significant (p [ 0.05)
which results in a large increase in osteoclastic resorption. This massive up-reg-

ulation in the removal of bone matrix is not matched by equivalent replacement of
bone therefore there is a net loss of bone mass. The increased number and depth of
resorption events results in perforation of individual trabeculae, particularly rod-
like structures, which now unloaded are targeted for further resorption and com-
pletely removed. This process results in disconnectivity in the structure locally and
as a whole and reduces the strength of the structure to a greater degree than
accounted for by the loss of bone mass alone [97, 107]. In addition, plate-like
trabeculae become perforated and over a number of cycles become more rod-like,
which in turn are susceptible to complete perforation and removal. While the
increased turnover in females returns to pre-menopausal levels after a number of
years, the transformation of the trabecular bone structural can be dramatic with an
associated increase in fracture risk.
In males, loss of bone mass from the 6th decade is more gradual than for
females and is associated with decreased androgen production [68, 92]. There is
increased bone activation of osteoclasts but not to the extent seen in menopausal
females, which while resulting in decreased bone mass is not associated with
resorption of sufficient depth to perforate trabeculae. Hence, there is generalized
thinning of trabeculae but the overall structure remains intact. However, in both
males and females there is an increased fracture incidence, which is not completely
explained by the loss of bone mass, which again suggests that the mechanical
integrity of the trabecular bone structure has been compromised [68].
The sex specific changes to trabecular bone structure in aging, described
above, are generalizations, and for individuals will be a combination of trabecular
perforation and trabecular thinning, to a greater or lesser degree [9]. However,

significant trabecular bone loss is likely to occur in all individuals over the age of
60 and the degree to which this bone loss contributes to an increase in fracture risk
is dependent on multiple factors, including but not exclusively, peak bone mass,
degree of sarcopenia, propensity to falls and sunlight exposure [45]. At the bone
level alone, skeletal site, is an important risk factor, where the amount of bone, the
trabecular bone architecture and the cortical bone morphological properties differs
markedly at different sites [4].
8 Summary
A great deal of the knowledge of trabecular bone structure has been elucidated
from quantitative methods developed for analysis of histological sections. From
the pioneering work of Harold Frost in the 1960s to the present, the complex
architecture of trabecular bone has been acknowledged as a three-dimensional
entity, which has been optimized for its primary function of ensuring the habitual
loads extant on the skeleton do not allow it to fracture. Despite the practical
difficulties of describing a 3D structure from 2D images, workers in this field
have developed and utilized powerful quantitative tools, collectively known
as bone histomorphometry. These tools have provided quantitative character-
ization of the dimensions of trabeculae, the spatial arrangement between
trabeculae the cellular dynamics at trabecular surfaces and the dynamics of bone
mineralization.
While ex vivo investigations utilizing histomorphometry have provided
comprehensive characterisation of trabecular bone structure and determined
skeletal variation and morphological properties these observations have provided
an understanding of why and how fractures can occur but not in whom they are
going to occur. The X-ray-based imaging tools available today promise to enable
in vivo study of individuals at equivalent spatial resolution to histology or ex
vivo micro-CT imaging. The suite of tools available for the analysis of trabec-
ular bone as a 3D structure has been significantly expanded with the develop-
ment of tools that can isolate individual trabecular elements, enabling the
morphology of these structures to be measured. Together with finite-element-
based analysis, apparent mechanical properties can be obtained at the level of
individual trabecular elements to fully characterise the ability of the structure to
resist loads under varying conditions.
Acknowledgments The authors acknowledge the staff of the Bone and Joint Research
Laboratory, SA Pathology for their skill in sample preparation and quantitative analyses and The
National Health and Medical Research Council, Australia for grant funding.
References
1. Aaron, J.E., Makins, N.B., et al.: The microanatomy of trabecular bone loss in normal aging
men and women. Clin. Orthop. Relat. Res. 215, 260–271 (1987)
2. Aaron, J.E., Shore, P.A., et al.: Trabecular architecture in women and men of similar bone
mass with and without vertebral fracture: II. Three-dimensional histology. Bone 27,
277–282 (2000)
3. Amling, M., Herden, S., et al.: Polyostotic heterogeneity of the spine in osteoporosis.
Quantitative analysis and three-dimensional morphology. Bone Miner. 27, 193–208 (1994)
4. Amling, M., Herden, S., et al.: Heterogeneity of the skeleton: comparison of the trabecular
microarchitecture of the spine, the iliac crest, the femur and the calcaneus. J. Bone Miner.
Res. 11, 36–45 (1996)
5. Anonymous. World Medical Association Declaration of Helsinki. W. M. Association (2008)
6. Badiei, A., Bottema, M.J., et al.: Influence of orthogonal overload on human vertebral
trabecular bone mechanical properties. J. Bone Miner. Res. 22, 1690–1699 (2007)
7. Bevill, G., Eswaran, S.K., et al.: Influence of bone volume fraction and architecture on
computed large-deformation failure mechanisms in human trabecular bone. Bone 39,
1218–1225 (2006)
8. Biewener, A.A.: Safety factors in bone strength. Calcif. Tissue Int. 53(Suppl 1), 68–74
(1993)
9. Birkenhager-Frenkel, D.H., Courpron, P., et al.: Age-related changes in cancellous bone
structure. Bone Miner. 4, 197–216 (1988)
10. Borah, B., Defresne, T.E., et al.: Long-term risedronate treatment normalizes mineralization
and continues to preserve trabecular architecture: sequential triple biopsy studies with
micro-computed tomography. Bone 39, 345–352 (2006)
11. Borah, B., Ritman E.L., et al.: The effect of risedronate on bone mineralization as measured by
micro-computed tomography with synchrotron radiation: correlation to histomorphometric
indices of turnover. Bone 37, 1–9 (2005)
12. Bouxsein, M.L., Boyd, S.K., et al.: Guidlines for assessment of bone microsctructure in
rodents using micro-computed tomography. J. Bone Miner. Res. 25, 1468–1486 (2010)
13. Burr, D.B.: Targeted and nontargeted remodeling. Bone 30, 2–4 (2002)
14. Byers, S., Moore, A., et al.: Quantitative histomorphometric analysis of the human growth
plate from birth to adolescence. Bone 27, 495–501 (2000)
15. Callis, G.M.: Bone. Theory and practice of histological techniques. In: Bancroft, J.D.,
Gamble, M. (eds.) pp. 333–364. Churchill Livingstone, London (2008)
16. Carneiro, R.M., Prebehalla, L., et al.: Lactation and bone turnover: a conundrum of marked
bone loss in the setting of coupled bone turnover. J. Clin. Endocrinol. Metab. 95, 1767–1776
(2010)
17. Chappard, C., Peyrin, F., et al.: Subchondral bone micro-architectural alterations
in osteoarthritis: a synchrotron micro-computed tomography study. Osteoarthr. Cartil.
14, 215–223 (2006)
18. Chappard, D., Legrand E., et al.: Measuring trabecular bone architecture by image analysis
of histological sections. Microsc. Anal. 13, 23–25 (1997)
19. Chappard, D., Legrand, E., et al.: Altered trabecular architecture induced by corticosteroids:
a histomorphometric study. J. Bone Miner. Res. 5, 676–685 (1996)
20. Chavassieux, P., Meunier, P.J.: Histomorphometric approach of bone loss in men. Calcif.
Tissue Intern. 69, 209–213 (2001)
21. Compston, J.E.: Bone histomorphometry––the renaissance. BoneKEy-Osteovis. 1, 9–12
(2004)
22. Compston, J.E., Mellish, R.W.E., et al.: Structural mechanisms of trabecular bone loss in
man. Bone Miner. 6, 339–350 (1989)
23. Currey, J.: Minimum mass of cancellous bone. Bone: structure and mechanics J. Currey,
pp. 224–225. Princeton, Princeton University Press (2002)
24. Dempster, D.W., Ferguson-Pell, M., et al.: Relationships between bone structure in the iliac
crest and bone structure in the lumbar spine. Osteoporos. Int. 3, 90–96 (1993)
25. Ebbesen, E.N., Thomsen, J.S., et al.: Age- and gender-related differences in vertebral bone
mass, density, and strength. J. Bone Miner. Res. 14, 1394–1403 (1999)
26. Eckstein, F., Matsuura, M., et al.: Sex differences of human trabecular bone microstructure
in aging are site specific. J. Bone Miner. Res. 22, 817–824 (2007)
27. Engelke, K., Gluer, C.C., et al.: Structural and fractal analyses of the trabecular network
using micro computed tomography images. J. Bone Miner. Res. 8, S354 (1993)
28. Fazzalari, N.L., Crisp, D.J., et al.: Mathematical modelling of trabecular bone structure: the
evaluation of analytical and quantified surface to volume relationships in the femoral head
and iliac crest. J. Biomech. 22, 901–910 (1989)
29. Fazzalari, N.L., Darracott, J., et al.: A quantitative description of selected stress regions of
cancellous bone in the head of the femur using automatic image analysis. Metab. Bone Dis.
Relat. Res. 5, 119–125 (1983)
30. Fazzalari, N.L., Forwood, M.R., et al.: Assessment of cancellous bone quality in severe
oseoarthritis: bone mineral dnesity, mechanics and microdamage. Bone 22, 381–388 (1998)
31. Fazzalari, N.L., Kuliwaba, J.S., et al.: The ratio of messenger RNA levels of receptor
activator of nuclear factor kB ligand to osteoprotegerin correlates with bone remodeling
indices in normal human cancellous bone but not in osteoarthritis. J. Bone Miner. Res. 16,
1015–1027 (2001)
32. Fazzalari, N.L., Kuliwaba, J.S., et al.: Cancellous bone microdamage in the proximal femur:
influence of age and osteoarthritis on damage morphology and regional distribution. Bone
31, 697–702 (2002)
33. Fazzalari, N.L., Moore, A., et al.: Quantitative analysis of trabecular morphogenesis in the
human costochondral junction during postnatal period in normal subjects. The Anat. Rec.
248, 1–12 (1997)
34. Fazzalari, N.L., Moore, R.J., et al.: Comparative study of iliac crest and subchondral
femoral bone in osteoarthritic patients. Bone 13, 331–335 (1992)
35. Fazzalari, N.L., Parkinson, I.H.: Femoral trabecular bone of osteoarthritic and normal
subjects in an age and sex matched group. Osteoarthr. Cartil. 6, 377–382 (1998)
36. Fazzalari, N.L., Parkinson, I.H., et al.: Antero-postero differences in cortical thickness and
cortical porosity of thoraco-lumbar vertebral bodies. Jt. Bone Spine 73, 293–297 (2006)
37. Feldkamp, L.A., Goldstein, S.A., et al.: The direct examination of three-dimensional bone
architecture in vitro by computed tomography. J. Bone Miner. Res. 4, 3–11 (1989)
38. Fields, A.J., Eswaran, S.K., et al.: Role of trabecular microarchitecture in whole-vertebral
body biomechanical behavior. J. Bone Miner. Res. 24, 1523–1530 (2009)
39. Frost, H.M.: Tetracycline-based histological analysis of bone remodeling. Calcif. Tissue
Res. 3, 211–237 (1969)
40. Frost, H.M.: Bone histomorphometry: analysis of trabecular bone dynamics. In: Recker R.R.
(ed.) Bone histomorphometry: techniques and interpretation, pp. 109–131. CRC Press, Boca
Raton (1983)
41. Frost, H.M.: From Wolff’s law to the Utah paradigm: insights about bone physiology and its
clinical applications. The Anatomical Record 262, 398–419 (2001)
42. Glorieux, F.H., Salle, B.L., et al.: Dynamic histomorphometric evaluation of human fetal
bone formation. Bone 12, 377–381 (1991)
43. Guggenbuhl, P.: Osteoporosis in males and females: is there really a difference? Jt. Bone
Spine 76, 595–601 (2009)
44. Guo, X.E., Kim, C.H.: Mechanical consequence of trabecular bone loss and its treatment: a
three-dimension model simulation. Bone 30, 404–411 (2002)
45. Heaney, R.P., Abrams, S., et al.: Peak bone mass. Osteoporos. Int. 11, 985–1009 (2000)
46. Hildebrand, T., Laib, A., et al.: Direct three-dimensional morphometric analysis of human
cancellous bone: microstrucural data from spine, femur, iliac crest, and calcaneus. J. Bone
Miner. Res. 14, 1167–1174 (1999)
47. Hildebrand, T., Ruegsegger, P.: A new method for the model-independent assessment of
thickness in three-dimensional images. J. Microsc. 185, 67–75 (1997)
48. Hildebrand, T., Ruegsegger, P.: Quantification of bone microarchitecture with the structure
model index. Comput. Meth. Biomech. Biomed. Eng. 1, 15–23 (1997)
49. Homminga, J., McCreadie, B.R., et al.: Cancellous bone mechanical properties from
normals and patients with hip fractures differ on the structural level, not on the bone hard
tissue level. Bone 30, 759–764 (2002)
50. Hordon, L.D., Raisa, M., et al.: Trabecular architecture in women and men of similar bone
mass with and without vertebral fracture: I. Two dimensional histology. Bone 27, 271–276
(2000)
51. Ito, M., Nakmura, T., et al.: Analysis of trabecular microarchitecture of human iliac bone
using microcomputed tomography in patients with hip arthrosis with and without vertebral
fracture. Bone 23, 163–169 (1998)
52. Johnell, O., Kanis, J.A.: An estimate of the worldwide prevalence, mortality and disability
associated with hip fracture. Osteoporos. Int. 15, 897–902 (2004)
53. Johnell, O., Kanis, J.A.: An estimate of the worldwide prevalence and disability with
osteoporotic fractures. Osteoporos. Int. 17, 1726–1733 (2006)
54. Kanis, J.A., Borgstrom, F., et al.: Assessment of fracture risk. Osteoporos. Int. 16, 581–589
(2005)
55. Kaufman, J.M., Goemaere, S.: Osteoporosis in men. Best Pract. Res. Clin. Endocrinol.
Metab. 22, 787–812 (2008)
56. Kimmel, D.B., Jee, W.S.S.: Measurements of area, perimeter and distance: details of data
collection in bone histomorphometry. Bone histomorphometry: techniques and
interpretation. Recker R.R., pp. 89–108. Boca Raton, CRC Press (1983)
57. Krug, R., Banerjee, S., et al.: Feasibility of in vivo structural analysis of high-resolution
magnetic resonance images of the proximal femur. Osteoporos. Int. 16, 1307–1314 (2005)
58. Kuliwaba, J.S., Findlay, D.M., et al.: Enhanced expression of oseocalcin mRNS in human
osteoarthitic trabecular bone of the proximal femur is associated with decreased expression
of interleukin-6 and interleukin-11 mRNA. J. Bone Miner. Res. 15, 332–341 (2000)
59. Ladinsky, G.A., Vasilic, B., et al.: Trabecular structure quantified with the MRI-based
virtual bone biopsy in postmenopausal women contributes to vertebral deformity burden
independent of areal vertebral BMD. J. Bone Miner. Res. 23, 64–74 (2008)
60. Landin, L.A.: Fracture patterns in children. Acta Orthop. Scand. 202, 100–109 (1983)
61. Link, T.M., Vieth, V., et al.: High-resolution MRI vs multislice CT: which technique depicts
the trabecular structure best? Eur. Radiol. 13, 663–671 (2003)
62. Liu, X.S., Bevill, G., et al.: Micromechanical analyses of vertebral trabecular bone based on
individual trabeculae segmentation of plates and rods. J. Biomech. 42, 249–256 (2009)
63. Liu, X.S., Sajda, P., et al.: Complete volumetric decomposition of individual trabecular
plates and rods and its morphological correlations with anisotropic elastic moduli in human
trabecular bone. J. Bone Miner. Res. 23, 223–235 (2008)
64. Liu, X.S., Sajda, P., et al.: Quantification of the roles of trabecular microarchitecture and
trabecular type in determining the elastic modulus of human trabecular bone. J. Bone Miner.
Res. 21, 1608–1617 (2006)
65. Liu, X.S., Zhang, X.H., et al.: Contributions of trabecular rods of various orientations in
determining the elastic properties of human vertebral trabecular bone. ASME Summer
Bioengineering Conference, Keystone, Colorado (2007)
66. Lochmuller, E.M., Matsuura, M., et al.: Site-specific deterioration of trabecular bone
architecture in men and women with advancing age. J. Bone Miner. Res. 23, 1964–1973
(2008)
67. Majumdar, S.: A review of magnetic resonance (MR) imaging of trabecular bone micro-
architecture: contribution to the prediction of biomechanical properties and fracture
prevalence. Technol. Health Care 6, 321–327 (1998)
68. Mazess, R.B.: On aging bone loss. Clin. Orthop. Relat. Res. 165, 239–252 (1982)
69. Meunier, P.J.: Histomorphometry of the skeleton. In: Peck, W.A. (ed.) Bone and mineral
research annual, pp. 191–222. Excerpta Medica, Amsterdam (1983)
70. Moore, R.J., Durbridge, T.C., et al.: Trabecular spacing in post-menopausal Australian
women with and without vertebral fractures. Australian and New Zealand J. Med. 22,
269–273 (1992)
71. Morgan, E.F., Keaveny, T.M.: Dependence of yoield strain of human trabecular bone on
anatomic site. J. Biomech. 34, 569–577 (2001)
72. Mori, S., Harruff, R., et al.: Trabecular bonevolume and microdamage accumulation in the
femoral heads of women with and without femoral neck fractures. Bone 21, 521–526 (1997)
73. Mosekilde, L.: Sex differences in age-related loss of vertebral trabecular bone mass and
structure-biomechanical consequences. Bone 10, 425–432 (1989)
74. Mueller, T.L., Van Lenthe, G.H., et al.: Regional, age and gender differences in architectural
measures of bone quality and their contribution to bone mechanical competence in the
human radius of an elderly population. Bone 45, 882–891 (2009)
75. Muller, R., Hahn, M., et al.: Morphometric analysis of noninvasively assessed bone
biopsies: comparison of high-resolution computed tomography and histological sections.
Bone 18, 215–220 (1996)
76. Muller, R., Hildebrand, T., et al.: Non-invasive bone biopsy: a new method to analyse and
display the three-dimensional structure of trabecular bone. Phys. Med. Biol. 39, 145–164
(1994)
77. Muller, R., Koller, B., et al.: Resolution dependency of microstructural properties of
cancellous bone based on three-dimensional micro-tomography. Technol. Health Care 4,
113–119 (1996)
78. Muller, R., van Campenhout, H., et al.: Morphometric analysis of human bone biopsies: a
quantitative structural comparison of histological sections and micro-computed
tomography. Bone 23, 59–66 (1998)
79. Odgaard, A.: Three-dimensional methods for quantification of cancellous bone architecture.
Bone 20, 315–328 (1997)
80. Odgaard, A., Gundersen, H.J.: Quantification of connectivity in cancellous bone, with
special emphasis on 3-D reconstructions. Bone 14, 173–182 (1993)
81. Pahr, D.H., Zysset, P.K.: A comparison of enhanced continuum FE with micro FE models of
human vertebral bodies. J. Biomech. 42, 455–462 (2009)
82. Parfitt, A.M.: Size of bone in the ages: endocortical resorption. J. Bone Miner. Res. 17, 1306
(2002)
83. Parfitt, A.M.: Targeted and nontargeted bone remodeling: Relationship to basic
multicellular unit origination and progression. Bone 30, 5–7 (2002)
84. Parfitt, A.M., Drezner, M.K., et al.: Bone histomorphometry: standardization of
nomenclature, symbols, and units. J. Bone Miner. Res. 2, 595–610 (1987)
85. Parfitt, A.M., Mathews, C.H.E., et al.: Relationships between surface, volume and thickness
if iliac trabecular bone in aging and in osteoporosis. J. Clin. Investig. 72, 1396–1409 (1983)
86. Parkinson, I.H., Fazzalari, N.L.: Cancellous bone structure analysis using image analysis.
Australasian Phys. Eng. Sci. Med. 417, 64–67 (1994)
87. Parkinson, I.H., Fazzalari, N.L.: Interrelationships between structural parameters of
cancellous bone reveal accelerated structural change at low bone volume. J. Bone Miner.
Res. 18, 2200–2205 (2003)
88. Parkinson, I.H., Forbes, D., et al.: Model-independent 3D descriptorss of vertebral
cancellous bone architecture. J. Osteoporos. (2010). doi:10.4061/2010/641578
89. Reginster, J.-Y., Minne, H.W., et al.: Randomized trial of the effects of risedronate on
vertebral fractures in women with established postmenopausal osteoporosis. Osteoporos.
Int. 11, 83–91 (2000)
90. Riggs, B.L., Melton, L.J.: Involutional osteoporosis. The New England J. Med. 314,
1676–1686 (1986)
91. Riggs, B.L., Melton, L.J.: Bone turnover matters: the raloxifene treatment paradox of
dramatic decreases in vertebral fractures without commensurate increases in bone density.
J. Bone Miner. Res. 17, 11–14 (2002)
92. Riggs, B.L., Melton 3rd, L.J., et al.: Population-based study of age and sex differences in
bone volumetric density, size, geometry, and structure at different skeletal sites. J. Bone
Miner. Res. 19, 1945–1954 (2004)
93. Robling, A., Castillo, A., et al.: Biomechanical and molecular regulation of bone
remodeling. Annu. Rev. Biomed. Eng. 8, 455–498 (2006)
94. Roux, J.P., Wegrzyn, J., et al.: Contribution of trabecular and cortical components to
biomechanical behaviour of human vertebrae: an ex vivo study. J. Bone Miner. Res. 25,
356–3561 (2010)
95. Schaffler, M.B., Choi, K., et al.: Aging and matrix microdamage accumulation in human
compact bone. Bone 17, 521–525 (1995)
96. Schuit, S.C.E., van der Klift, M., et al.: Fracture incidence and association with bone
mineral density in elderly men and women: the Rotterdam study. Bone 34, 195–202 (2004)
97. Silva, M.J., Gibson, L.J.: Modeling the mechanical behavior of vertebral trabecular bone:
effects of age-related changes in microstructure. Bone 21, 191–199 (1997)
98. Simpson, E.K., Parkinson, I.H., et al.: Intervertebral disc disorganisation is related to
trabecular bone architecture in the lumbar spine. J. Bone Miner. Res. 16, 681–687 (2001)
99. Sornay-Rendu, E., Boutroy, S., et al.: Cortical and trabevular architecture are altered in
postmenopausal women with fractures. Osteoporos. Int. 20, 1291–1297 (2009)
100. Stauber, M., Muller, R.: Age-related changes in trabecular bone microstructures: global and
local morphometry. Osteoporos. Int. 17, 616–626 (2006)
101. Stauber, M., Muller, R.: Volumetric spatial decomposition of trabecular bone into rods and
plates—a new method for local bone morphometry. Bone 38, 475–484 (2006)
102. Stauber, M., Rapillard, L., et al.: Importance of individual rods and plates in the assessment
of bone quality and their contribution to the bone stiffness. J. Bone Miner. Res. 21, 586–595
(2006)
103. Szulc, P., Kaufman, J.M., et al.: Biochemical assessment of bone turnover in men.
Osteoporos. Int. 18, 1451–1461 (2007)
104. Thomsen, J.S., Ebbesen, E.N., et al.: Static histomorphometry of human iliac crest and
vertebral trabecular bone: a comparative study. Bone 30, 267–274 (2002)
105. Townsend, P.R., Rose, R.M., et al.: Buckling studies of single human trabeculae.
J. Biomech. 8, 199–201 (1975)
106. Tsangari, H., Findlay, D.M., et al.: Structural and remodeling indices in the cancellous bone
of the proximal femur across adulthood. Bone 40(1), 211–217 (2006)
107. van der Linden, J.C., Homminga, J., et al.: Mechanical consequences of bone loss in
cancellous bone. J. Bone Miner. Res. 16, 457–465 (2001)
108. van Staa, T.P., Dennison, E.M., et al.: Epidemiology of fractures in England and Wales.
Bone 29, 517–522 (2001)
109. Wang, Q., Seeman, E.: Skeletal growth and peak bone strength. Best Pract. Res. Clin.
Endocrinol. Metab. 22, 687–700 (2008)
110. Wegrzyn, J., Roux, J.P., et al.: Role of trabecular microarchitecture and its heterogeneity
parameters in the mechanical behavior of ex vivo human L3 vertebrae. J. Bone Miner. Res.
25, 2324–2331 (2010)
111. Weinstein, R.S., Hutson, M.S.: Decreased trabecular width and increased trabecular spacing
contribute to bone loss with aging. Bone 8, 137–142 (1987)
112. Whitehouse, W.J.: The quantitative morphology of anisotropic trabecular bone. J. Microsc.
101, 153–168 (1974)
113. Yeni, Y.N., Zinno M.J., et al.: Variability of trabecular microstructure is age-, gender-, race-
and anatomic site-dependent and affects stiffness and stress distribution properties of human
vertebral cancellous bone. Bone 49, 886–894 (2011)
Cortical Bone Mechanics and Composition:
Effects of Age and Gender
Xiaodu Wang
Abstract Bone fragility fractures are a major health care concern for postmenopausal
women and the elderly of both genders. Postmenopausal and age-related osteoporosis/
osteopenia is a major contributor to the risk of such fractures. Since cortical bone is
the major load bearing tissue, the effects of age, gender, and pathological changes on
the mechanical competence of cortical bone tissues have been of great interest to bone
researchers. This chapter provides the information on the current understanding of
the micro/ultrastructural and compositional properties and their contribution to the
bulk mechanical (elastic, plastic, and viscous) behavior of cortical bone tissues.
In addition, the effect of age and gender on the structural/compositional properties and
their impacts on the mechanical competence of cortical bone are also discussed.

Keywords Cortical bone Elasticity Plasticity Bone strength Aging Gender
1 Introduction
Structurally, bone functions as a load bearing tissue to support and protect the
human body for daily physical activities. Bone can be classified into two types:
cortical and trabecular. Cortical bone is a dense tissue that makes up about 80% of
the mass of the skeleton, and is found primarily in the shafts of long bones, the
outer shell at the ends of joints, the vertebrae and ribs. Trabecular bone has a
porous structure and is located in medullary cavities at the ends of long bones and
the interior of short bones such as ribs and vertebrae [1].
X. Wang (&)
Mechanical and Biomedical Engineering, University of Texas at San Antonio,
One UTSA Circlet, San Antonio, TX 78249, USA
e-mail: xiaodu.wang@utsa.edu

DOI: 10.1007/8415_2011_108
Published Online: 6 December 2011
54 X. Wang
Bone fractures are a major health care concern for postmenopausal women and the
elderly of both genders. Postmenopausal and age-related osteoporosis/osteopenia is
a major contributor to the risk of such fractures. Since cortical bone is the major load
bearing tissue, age-related and pathological changes in cortical bone tissues have
been of interest to bone researchers. Although there are multiple factors, cortical
bone fragility is ultimately determined by two biomechanical factors: its micro-
structure and intrinsic tissue quality. The microstructural integrity is mainly related
to porosity and some other microstructural features (e.g., osteon/interstitial, cement
lines), whereas the intrinsic tissue quality is primarily affected by the composition
and ultrastructural features of bone constituents (i.e., mineral, collagen, and water).
This chapter intends to provide a fundamental review regarding the composition and
mechanical behavior of cortical bone and how these are affected by aging and gender.
2 Hierarchy of Human Cortical Bone
Human cortical bone has a complex, hierarchical structure. At its most simple
form it can be considered a two-phase material (*95% solid, *5% porous).
A typical section of a human long bone reveals the cortical microstructure of
osteons and interstitial tissue between periosteal and endosteal lamellae (Fig. 1).
Lamellae are layered sheet-like structures and serve as basic building units of
human cortical bone. Osteons (also called Haversian systems) are tube-like
structures of multiple concentric lamellae with a canal in the center that accom-
modates blood vessels for transportation of nutrients and removal of wastes by
bone cells (e.g., osteocytes). The interstitial tissue is also lamellar by nature and is
actually the remnants of either primary bone or previously formed osteons that are
partially removed by the bone remodeling process.
2.1 Composition of Cortical Bone
Solid bone tissue can be characterized as a natural composite of mineral (apatite

crystals), organic matrix (mostly type I collagen fibrils), and water [2–5]. The
mineral phase comprises *60% by weight (*40% by volume), the organic matrix
*30% by weight (*40% by volume), and water *10% by weight (*25% by
volume) [6, 7].
2.1.1 Mineral Phase
The mineral phase mainly consists of crystals that have a composition of calcium
(Ca2+) and phosphate (PO43-) with a small fraction of carbonates (CO32-) and
other ‘‘impurities’’ (sodium, magnesium, potassium, citrate, fluoride, HPO3-) [8].
Cortical Bone Mechanics and Composition: Effects of Age and Gender 55
Osteonal Periosteal
lamellae lamellae
Osteon
Interstitial
tissue
Cortical
bone
Endosteal Haversian
lamellae canal
Fig. 1 Hierarchical structure of cortical bone that is composed of lamellae, osteon, and
interstitial tissues
The mineral crystals in bone are plate-like in shape and quite small, having the
length, width, and thickness of 50 9 25 nm 9 *1.5–4.0 nm, respectively [9–11].
Early X-ray diffraction studies indicate that the bone mineral is similar to
hydroxyapatite, Ca5(PO4)3(OH) [12, 13], but with some distinctions compared
to synthetic hydroxyapatite [14]. Such distinctions are usually considered due to
non-stoichiometric ratio of calcium to phosphorous, presence of strongly bound
water, and deposition of amorphous mineral (tricalcium or octacalcium phosphate)
[8]. Some studies suggest that the mineral phase may be classified as a carbonated
apatite Ca5(PO4CO3)3 since hydroxyl groups are not observed when bone is
analyzed with Fourier Transform Infrared Spectroscopy (FTIR) or Nuclear
Magnetic Resonance (NMR) [15, 16].
2.1.2 Organic Matrix
The organic matrix consists of collagen and non-collagenous proteins, with type I
collagen being the major part ([90%). Secreted by bone forming cells known as
osteoblasts, procollagen is a triple helical rod of three intertwining polypeptide
chains (two identical a1 helices and one different a2 helix), each containing
approximately 1,000 amino acids, in which every third residue is glycine and
positioned toward the center of the super-coil [17]. Proline typically occupies the
next position, and there is an abundance of hydroxyproline in the third position.
Hydroxylysine is a unique residue of bone collagen and gives rise to cross-linking.
Upon enzymatic cleavage of the non-helical, amino (N) and carboxyl (C)
terminals, procollagen forms collagen fibrils by self-assembling via crosslinks
into a staggered arrangement. In general, collagen fibrils are 30–80 nm in
diameter [18]. There are no data currently available regarding the length distri-
bution of collagen fibrils in bone. However, it is anticipated that they are longer
56 X. Wang
than a few micrometers. The collagen crosslinks not only determine fibril
arrangement, but also affect the later mineralization process. The enzymatic
control of crosslinks occurs through the lysyl and hydroxylysyl residues at both C-
and N-terminal ends of collagen molecule [19]. In addition, pyrrole (another non-
reducible, enzymatic crosslink) is formed when hydroxylysyl aldehyde reacts with
non-hydroxylated lysine [20]. There are more divalent than trivalent crosslinks at
the beginning of collagen fibril formation but the discrepancy decreases as the
skeleton matures [21]. Collagen crosslinks may also form through age-related non-
enzymatic pathways, namely glycation. Briefly, glucoses react with certain amino
groups (e.g., lysine and arginine) of long-lived proteins, resulting in a rearrange-
ment of aldimine linkages into more stable keto-imine linkages and producing so
called Amadori products [22]. Further oxidative breakdown occurs over time
(tissue aging), thus causing reactions with other amino acid residues to form
advanced glycation end-products (AGEs) [23]. The collagen crosslinks induced by
AGEs have shown a significant correlation with changes in the toughness of bone
[24, 25].
Numerous non-collagenous proteins found in bone may influence the recruit-
ment, attachment and differentiation of bone cells, and also the structural integrity
of the tissue [26]. The most abundant non-collagenous protein is osteocalcin,
which is produced by osteoblasts and believed to be related to bone calcification
[27–29]. As structural proteins, non-collagenous proteins may also contribute to
bone mechanical integrity including strength, hardness and flexibility [30, 31].
2.1.3 Water
Water is distributed throughout bone in three forms: freely mobile in vascular-

lacunar-canalicular space, bound to the surface of the collagen and mineral phase,
and solid-like within the collagen and mineral molecules [32–34]. Water (H2O)
molecules are polar in nature, having a more negative charge on the oxygen side
and a more positive charge on the hydrogen side. Thus, H2O naturally associates
itself with mineral (PO4- or Ca2+) and collagen (glycine, hydroxyproline, car-
boxyl, and hydroxylysine). Water associates with collagen at two levels: the
hydrogen bonds within the triple helix of collagen molecules (due to the hydroxyl
group of hydroxyproline) and the hydrogen bonds with the polar side chains of
collagen fibrils [35, 36].
2.2 Microstructure of Cortical Bone
Osteons and interstitial tissues in human cortical bone are formed by lamellae [1].
A lamella is a sheet-like structure of *3–5 lm thick and resembles a fiber rein-
forced composite material, in which collagen fibrils act as the reinforcement phase
and mineral crystals as the matrix (Fig. 2). In general, collagen fibrils in lamellae
Fig. 2 Lamella can be Cortical bone Osteon Lamella

considered as a sheet of long Osteon
fiber reinforced composite
Interstitial
Fig. 3 Collagen fibril

distributions in a lamella
of bone
have a preferred but shifting orientation (Fig. 3). With reflective light microscopy,
lamellae in osteons appear as white bands of varying thickness separated by a thin
dark layer, which results from the difference in the orientation of collagen fibrils
between neighboring lamella. Two general architectures of lamellae are postu-
lated: (1) ‘‘orthogonal plywood’’ with alternating orthogonal orientations of fibrils,
and (2) ‘‘twisted plywood’’ with continually changing orientation of fibrils in
which the pattern repeats itself through 180° cycles [37]. TEM and SEM obser-
vations also show that parallel fibrils may rotate at a plywood angle of *30°
through several successive sub-layers of varying thickness in a lamella [38, 39].
The fibrils may intermingle across lamellae, but there is a distinct and preferred
orientation for any given layer [5].
2.3 Ultrastructure of Cortical Bone
The nano-scale structure (often called ultrastructure) and the interactions between
mineral, collagen and water in bone are still poorly understood. For example, some
investigators argue that most mineral crystals reside in the intrafibrillar spaces
(e.g., gap regions) during mineralization of collagen fibrils [40, 41]. However,
some studies indicate that only limited portion of the mineral phase is in the
intrafibrillar space, whereas a large percentage of mineral crystals are deposited
outside of collagen fibrils [42, 43].
58 X. Wang
Fig. 4 Molecular packing

models of mineralized
collagen fibrils and estimated
mechanical properties
presented by Jäger and
Fratzl [46]
Transmission electron micrographs of individual mineralized collagen fibrils

show that hydroxyapatite crystals are located mainly at the level of the gap regions
within the fibrils and more or less uniformly stacked across the fibril diameter [40].
In addition, high-voltage-electron stereomicroscopy reveals that bone mineral
crystals are located within collagen fibrils, suggesting that there exists a local
‘‘bulging’’ along the fibrils corresponding to a 68 nm periodicity, which contains
additional mass of minerals [44]. Moreover, atomic force microscopy (AFM)
analysis verifies that the gap regions in collagen fibrils are indeed filled with mineral
crystals [45]. In addition to the experimental observations, there are several attempts
in modeling mineralized collagen fibrils [41, 46]. For example, a model with a
staggered array of mineral crystal platelets embedded in a collagen matrix (Fig. 4) is
proposed to predict the molecular packing in collagen fibrils, which can be used to
predict both elastic modulus and fracture stress as a function of the amount of mineral
in the fibril [46]. Another model of mineralized collagen fibrils based on the data of
neutron diffraction, electron microscopy, crosslinking, and composition-density
predicts that three quarters of the mineral in bone is disposed within the fibrils [41].
Beyond the mineralized collagen fibril level, several models are proposed to
describe the interaction between the fibrils. Early studies propose a simple fiber-
reinforced composite to model the elastic properties of osteonal bone, assuming
(1) bone collagen fibrils are not principally aligned along the long axis of the bone,
but demonstrate an alignment of 30° with respect to the long axis, (2) 75% of
mineral crystals reside outside of collagen fibrils, and (3) mineral crystals outside
of collagen fibrils have their c-axis in the longitudinal direction [47]. Recently,
a multiscale modeling of bone as a fiber reinforced composite material is
proposed to describe the elastic properties of bone considering both mineralized
collagen fibrils and the extrafibrillar minerals, which are mechanically equivalent
Fig. 5 Schematic
representation of glue like
bonds between mineralized
collagen fibrils presented by
Fantner et al. [49]
to reinforcing rings coated on each individual fibril. In this model, no more than
30% of the total mineral content is extrafibrillar and the fraction of extrafibrillar
minerals grows linearly with the overall degree of mineralization. The model
predictions for the elastic moduli and constants are found to be in a good agree-
ment with the experimental data reported in the literature [48].
Recently, a so-called glue-like bond model is proposed to describe the inter-
action between the mineralized collagen fibrils [49]. In this model, the mineralized
collagen fibrils are held together by a non-fibrillar organic matrix (Fig. 5), which
acts as a glue. Such glue resists both separation and slippage between the min-
eralized collagen fibrils, thereby helping transmit force between fibrils. In addition,
so-called sacrificial bonds are formed in the organic matrix of the glue. When the
matrix is stretched, energy is dissipated through rupturing of sacrificial bonds and
by stretching of molecules in the glue to release so-called hidden length. Since the
sacrificial bonds are reformable, the break and reform of such bonds further
increase the total energy dissipation of the tissue during the deformation of bone.
3 Mechanical Behavior of Human Cortical Bone
3.1 Mechanical Testing of Cortical Bone
Tension, compression, and torsion tests are mechanical tests commonly used to
evaluate mechanical properties of human cortical bone. In general, these tests are
performed in a monotonic manner. However, cyclic diagnostic tests are also
performed in order to obtain more information about the changes in bone
mechanical properties during the loading process [50–52].
3.1.1 Monotonic Tests
From monotonic tests, the uniaxial and torsional elastic modulus, yield stress/
strain, ultimate stress, failure strain, and toughness can be measured (Fig. 6).
60 X. Wang
Fig. 6 Typical stress–strain

curve of cortical bone
specimens in tension X
Stress
UT
0 0.2%
Strain
The elastic modulus (E) is estimated as the slope of the initial linear part of the
stress–strain curve. The yield stress and strain (ry and ey) are estimated using a
so-called 0.2% strain offset method. The ultimate stress (strength) of bone is
determined as the maximum stress from the stress–strain curve. The failure strain
or ductility is measured as the strain at failure of the specimen. The toughness (UT)
is estimated as the area under the stress–strain curve.
3.1.2 Cyclic Diagnostic Tests
Cyclic diagnostic tests allow for determining the changes in mechanical properties
of bone as a function of loading history [50–52]. Unlike the monotonic test, this
test subjects a specimen to multiple diagnostic cycles of a loading-dwell-unload-
ing-dwell-reloading sequence, in which the strain applied increases incrementally
with each cycle and dwell intervals account for bone’s viscoelastic behavior
(Fig. 7). In each cycle, the specimen is loaded at a constant loading rate to a
designated strain level (ei), held at that strain level for full stress relaxation (stress
relaxation dwell), unloaded with the same rate to zero stress, held for the full
recovery of creep strain (creep strain dwell), and finally reloaded to the next strain
level with a prescribed increment. This procedure is repeated until failure. During
the stress relaxation dwell, the viscoelastic response of the bone specimens can be
evaluated using the stress versus time curve (Fig. 7).
The mechanical properties that can be estimated at each diagnostic cycle are as
follows: (1) the instantaneous modulus (Ei) is the slope between two points, one at
the end of stress relaxation dwelling and the other at the end of creep strain
dwelling; (2) the maximum stress (rmaxi) is the peak stress in the cycle; (3) the
permanent strain (epi) is the residual strain value after full recovery of the creep
strain at the end of the dwelling period; (4) the total stress relaxation (Dri) is
defined as the total change in stress from the beginning to the end of the stress
relaxation dwelling period. From the curve, the energy dissipation can also be
determined. Briefly, released elastic strain energy (Ueri) is the surface energy
Previous diagnostic maxi

cycles Stress
relaxation
dwell maxi
Reload
Stress
Load
Stress relaxation
Uhi 0i
Umi Ei
Ueri Ue0
pi
Unload
Creep strain εi Time

dwell
Strain
Fig. 7 A novel progressive loading protocol designed to determine the evolution of bone
properties as a function of increasing strain
dissipation of newly formed microdamage, which can be estimated as the area

under the unloading curve minus the elastic recovery energy (Ue0i). Plastic strain
energy (Upi) is the energy dissipation by permanent deformation of bone, which
can be approximated as the summation of cumulative areas between loading
and reloading curves of the successive cycles (Umi) minus Ueri. Finally, hysteresis
energy (Uhi) is the energy dissipation by the viscoelastic response of bone,
which can be calculated as the area between the loading and unloading curves
(i.e., the hysteresis loop).
3.2 Elastic Properties of Cortical Bone
3.2.1 Elastic Modulus
The elastic properties of human cortical bone at the macroscopic (also called
apparent or continuum) level are usually considered transversely isotropic, with
the first principal direction along the longitudinal axis of the bone [53]. This is
largely because osteons are oriented along the long axis and located in a random
distribution in the transverse plane of bone. The experimental data (Table 1) show
that the elastic modulus of human cortical bone is much lower in the transverse
direction than the longitudinal direction, whereas the elastic moduli of bone in
transverse directions (transverse to the long axis) are similar [54, 55]. Thus, a
complete characterization of the elastic behavior of cortical bone requires five
elastic constants: longitudinal elastic modulus (EL) and Poisson’s ratio (mL),
transverse elastic modulus (ET) and Poisson’s ratio (mT), and transverse shear
modulus (GT).
62 X. Wang
Table 1 Elastic modulus of human cortical bone measured in monotonic tests

Orientation Bone Test Modulus (GPa) References
Longitudinal Femur Tension 15.6–19.4 [59, 60, 62, 148]
Compression 15.2–18.1 [60, 62, 149]
Torsion 3.3–5.0 [51, 54]
3pt-Bending 10.8–15.8 [63–66]
4pt-Bending 12.1 [150]
Tibia Tension 18.0–29.2 [59, 60]
Compression 25.9–35.3 [60]
Cantilever bending 10.6 [151]
Fibula Tension 18.5 [59]
Humerus Tension 17.2 [59]
Radius Tension 18.5 [59, 152]
Ulna Tension 18.4 [59]
Compression 14.2 [55]
Radial Ulna Compression 3.8 [55]
Lateral Compression 4.2 [55]
The elastic modulus of cortical bone depends strongly on its porosity [56].
Porosity also strongly influences the apparent, volumetric bone mineral density of
cortical bone, so measures of density strongly correlate with elastic modulus
[57, 58]. In addition, modulus may vary with anatomic location, loading mode,
orientation, degree of mineralization, and specimen size. As shown in Table 1,
samples from human tibias tend to have a higher elastic modulus than samples
from other sites including the femur; samples from the fibula, humerus, radius and
ulna were reported to have similar tensile elastic moduli [59, 60]. The elastic
properties may also change with anatomic locations within long bones [53]. For
instance, the elastic modulus of human cortical bone from male donors is greater
for femoral diaphysis than metaphysis [61].
Although some data suggest a difference between tensile and compressive
moduli, Reilly et al. [62] concluded that there was no significant difference
between these two loading modes based on paired comparisons from *200
samples of human femoral cortical bone. On the other hand, the flexural moduli
obtained from three-point and four-point bending tests are significantly different
(lower) from those of axial (tension or compression) tests [63–66]. It should be
noted that unlike a uniaxial test, the elastic modulus value obtained from bending
tests is based upon linear elastic beam theory, which may not fully reflect the true
elastic behavior of bone in bending.
Due to spatial heterogeneity and hierarchical features, the elastic behavior of
bone depends on specimen size (Table 2). Micro-specimens taken from osteons
and interstitial tissues of human femurs have revealed significantly different
properties when compared to those of macro-specimens. The tensile [67] and
compressive [68] moduli at the microscopic level are much lower than those
obtained at the macroscopic level. In contrast, torsion tests at the microscopic level
have obtained modulus values significantly higher than those obtained at the
Table 2 Elastic properties measured in monotonic tests using micro human cortical bone
specimens
Bone Test Modulus (GPa) Tissue type References
Femur Tension 3.9–11.7 Osteons [67]
Compression 3.3–9.3 Osteons [68]
Torsion 17.2–23.2 Osteons [69]
3pt-Bending 2.3–2.7 Osteons [153]
Tibia 3-pt-Bending 5.4 Micro-beams [154]
4pt-Bending 6.8 Micro-beams [155]
Fig. 8 Modulus loss of

human cortical bone in both
tension and compression
along the long axis of middle
aged human cadaveric tibias
macroscopic level [69]. Further, within each testing mode, scatter of elastic
modulus values is evident and most likely attributable to the varied degree of
mineralization of the osteons that have different biological ages, and the alter-
nating orientation of collagen fibrils in the osteons. Finally, nanoindentation
experiments produced measures of lamellar elastic moduli for human cortical bone
(average value of 17.7 ± 4.0 GPa for osteons and 19.3 ± 4.7 GPa for interstitial
bone tissue) [70].
3.2.2 Modulus (Stiffness) Loss
A decline (loss) of elastic modulus (stiffness) is well documented in fatigue tests

of bone, in which damage accumulates with increasing cycle number [71–75].
In addition, recent studies using a progressive loading scheme indicate that the
elastic modulus of human cortical bone decreases exponentially with the applied
strain for both tension and compression (Fig. 8). In general, it is well accepted that
the modulus loss is due to micro damage accumulation in bone, which is largely
64 X. Wang
Fig. 9 Released elastic

energy by damage
accumulation with respect to
the applied strain for human
cortical bone in both tension
and compression modes
dependent on the loading mode [76]. Cross-hatch linear cracks are often observed
in compression, whereas so-called diffuse damage most likely occur in tension
[77, 78]. Differences in micro damage accumulation between tension and com-
pression are also reflected in the distinct way of energy dissipation. The released
elastic strain energy (Uer) by formation of newly formed damage surfaces is less in
compression than in tension (Fig. 9). From the damage mechanics point of view,
the energy dissipation by micro damage accumulation is realized through the
formation of new crack surfaces (i.e., surface energy), thereby resulting in
decreased elastic modulus. It is speculated that crosshatched damage formation
causes minimal energy dissipation through creation of new crack surfaces
(i.e., released elastic strain energy) compared with the presumably less hazardous
diffuse damages in tension. Although the surface energy dissipation by micro crack
accumulation serves as one of toughening mechanisms of bone, its contribution to
the total deformation of bone is limited. The major mechanism for energy dissi-
pation still ties with the plastic deformation (residual strain) during the post-yield
behavior of bone.
3.3 Post-Yield Properties of Cortical Bone
3.3.1 Yielding
Yield stress and strain are usually determined using the conventional 0.2% strain
offset method. However, using the progressive diagnostic test, the yield strain can
be estimated more accurately [79]. In general, the values obtained by the 0.2%
strain offset method are usually greater than those determined using the progres-
sive diagnostic technique. The longitudinal yield strain of human femur estimated
using the 0.2% strain offset method is *0.6–0.7% in tension [79] and *0.6–1.1%
Table 3 Yield stress/strain of femoral cortical bone

Loading mode Yield strain (%) Yield stress (MPa) Type of test References
Tension 0.6–0.7 Monotonic [79]
Compression 0.6–1.1 Monotonic [80, 81]
Tension 0.4–0.5 Progressive [50]
Compression 0.7–0.8 Progressive [32, 82]
Tension 114 ± 3.1 Monotonic [54]
Torsion 0.13 ± 0.1 55.8 ± 3.8 Monotonic [156]
3pt-Bending 154 ± 13 Monotonic [65]
3pt-Bending 166 ± 12 Monotonic [117]
Fig. 10 Plastic deformation

(strain) of human cortical
bone (Tibia) with respect to
the applied strain in both
tension and compression
modes
in compression [80, 81] (Table 3). By plotting the permanent strain obtained using
the progressive diagnostic scheme versus the applied strain (Fig. 10) [50], the
yield strain is estimated to be *0.4–0.5% in tension and *0.7–0.8% in
compression [32, 82]. In addition, the plastic deformation is almost linearly
proportional to the applied strain.
3.3.2 Plastic Energy Dissipation
Plastic strain energy dissipation (Up) with respect to the applied strain is more than
two times greater in compression than in tension (Fig. 11) [82–84]. The contrib-
uting factors to the much higher plastic energy dissipation in compression are: (1)
higher stress is needed in compression to produce a similar (post-yield) strain
compared to that required in tension, and (2) bone can sustain greater plastic
deformation in compression than tension at the same applied strain level. It seems
likely that the predominant damage formed by compression (cross-hatch type [77])
allows for more plastic deformation than the damage formed by tension (diffuse
type [77]).
66 X. Wang
Fig. 11 Plastic strain energy

dissipation of human cortical
bone with respect to the
applied strain in both tension
and compression modes
Table 4 Ultimate strength and strain of human cortical bone measured in monotonic tests
Bone Test Orientation Strength (MPa) Ultimate strain (%) References
Tension Longitudinal 120–161 2.2–3.4 [54, 59, 60, 148]
Tension Transverse 53 0.7 [54]
Compression Longitudinal 150.3–209 1.7 [54, 59, 60, 149]
Femur Compression Transverse 131 [54]
Torsion Longitudinal 74.1 5.2 [51]
3pt-Bending 183.4–194 [63, 64, 157]
4pt-Bending 174 [150]
Tibia Tension 140.3–161 2.3–4.0 [59, 60]
Compression 158.9–213 [59, 60]
Fibula Tension 146.1 [59]
Humerus Tension 122.6 [59]
Radius Tension 149.1 [59]
Ulna Tension 148.1 [59]
3.3.3 Ultimate Strength and Strain
There is wide variability in failure strength and strain reported for human cortical
bones from different anatomic sites (Table 4). In contrast to modulus, bone
strength depends on loading direction: bone is *50% stronger in compression
than tension [60, 62]. Values of ultimate strength (specimens oriented longitudi-
nally) range from 120 to 161 MPa in tension, 120–213 MPa in compression,
174–194 MPa in bending, and 74 MPa in torsion. Bone strength is significantly
dependent on specimen orientation relative to loading direction. In general,
strength is much weaker in transverse direction compared to longitudinal.

In addition, the longitudinal ultimate strain is between 2.2 and 4.0% in tension,
*1.7% in compression, and *5.2% in torsion, whereas the transverse ultimate
strain in tension is only *0.7%.
3.3.4 Toughness
Toughness is usually evaluated by the total energy (work) required to produce

failure during a mechanical test, which is equivalent to energy dissipated by the
specimen until failure. Bone toughness depends on the level of micro damage
induced during deformation and the energy required to generate this micro damage
[85]. There are two general types of energy dissipation in bone: surface energy
dissipated by the newly formed surfaces, and plastic strain energy due to the
permanent deformation of the tissue. Since mineral and collagen phases each have
limited capacities to be plastically deformed, the plastic (residual) strain during the
post-yield deformation of bone is most likely due to the relative displacement
between the two phases [86].
Vashishth et al. [87] investigated the question of whether micro cracking during
loading is a toughening mechanism in bone, and reported that damage accumu-
lation in bone increases with crack extension thus leading to an increase in fracture
toughness of the tissue. Based on observations from ceramics and bone fracture
specimens, it is speculated that micro crack formation occurs in two stages
[87, 88]: the formation of a frontal process zone (Stage I) and the formation of a
process zone wake (Stage II). The formation of the microcracks absorbs energy,
thereby decelerating the crack propagation. This has been verified using laser
confocal microscopy in other studies [89], in which diffuse damage was found to
consistently form and accumulate in the bone matrix between canaliculi and in the
vicinity of lacunae. Besides crack-tip shielding, there are other possible tough-
ening mechanisms by damage accumulation. At the microstructure level, cement
lines are thought to deflect the crack path [90]. At the ultrastructure level, collagen
fibers can bridge the crack, thus retarding its growth [91].
3.4 Viscoelastic Properties of Cortical Bone
Cortical bone is viscoelastic. Two possible causes of bone viscoelasticity are: (1)
fluid flow within the bone porosity, and (2) viscous response of solid bone tissue.
The first mechanism is supported by the fact that bone viscosity depends on
hydration condition [92]. Wet bone exhibits a larger viscoelastic damping (tand)
than dry bone over a broad range of frequency (5 mHz–5 kHz in bending). The
viscoelastic relaxation due to fluid flow in bone usually occurs at fairly high
frequencies, perhaps above 10 kHz [93]. The value of tand in a frequency range of
1–100 Hz is relatively minimal, suggesting that fluid flow in bone might be
68 X. Wang
Fig. 12 Viscoelastic responses of bone with respect to the applied strain: a Total stress
relaxation of bone in both tension and compression. b Time constants for both tension and
compression obtained from the stress-time curves in the progressive tests
unimportant at low frequency [94]. In addition, temperature-scan tests of bone

specimens with different concentration of denatured collagens indicate that the
hydration condition, not the collagen phase, is significantly related to the behavior
of tand [92]. Moreover, damping increases with accumulation of mechanical
damage, in terms of increases in tand and frequency sensitivity of storage modulus
(E’) of bone [95, 96].
Viscoelastic behavior of bone changes with applied strain. As bone is contin-
uously deformed, stress relaxation increases up to an equilibrium state after
yielding (Fig. 12). Total stress relaxation (Dr, a measure of the viscous contri-
bution to total load bearing) is much greater in compression than tension, and the
time constant (s, inversely related to damping capability) is slightly greater in
compression. Since hysteresis energy dissipation is viscoelastic in nature, it is most
likely dependent on the magnitude of both Dr and s. Thus, a positive effect on the
hysteresis energy dissipation due to an increase in Dr could be cancelled out by a
negative effect induced by a decrease in damping property (s) [84]. This could
explain the fact that only a slight difference in hysteresis energy dissipation
between compression and tension is observed (Fig. 13).
In general, an exponential relaxation function (e-t/s, the Debye model) is often
used to describe stress relaxation behavior of viscoelastic materials. However, this
model is not suitable for bone. A linear combination of the Kohlrausch–Williams–
Watts (KWW) function and the Debye function can be employed to predict the
viscous response of bone [97, 98]:
b
r ¼ A1 eðt=s1 Þ þ A2 et=s2 þ C ð1Þ
where, A1 and A2 represent the contribution to the total stress relaxation, and s1 and
s2 are the time constants for the KWW and Debye processes, respectively, b is a
material constant that determines the acuteness of the KWW stress relaxation, and
C is a constant representing the stress component by the pure elastic deformation.
Fig. 13 Hysteresis energy

dissipation of bone as a
function of applied strain in
both tension and compression
The KWW relaxation process is believed to relate to the ultrastructural disorder

of collagen fibrils, whereas the Debye process is more associated with the
microstructural and inelastic (creep) behavior of bone [97, 98].
4 Effects of Aging on Mechanical Properties

of Human Cortical Bone
4.1 Age-Related Microstructural and Compositional Changes
It is well documented that aging causes deleterious changes in cortical bone, which
may consequently contribute to fragility fractures. The most prominent age-related
change in cortical bone at the microstructural level is the increased porosity, which
has been noted in numerous studies. Increased porosity contributes to decreased
bone density and is directly correlated to the decreased strength. In addition,
age-related alterations in the nanoscale and chemical properties of the solid bone
tissue (‘‘tissue quality’’) are another significant factor that directly affects bone
strength [32, 99–101]. One example is age-related accumulation of advanced
glycation end products (AGEs) that may alter the functionality of the collagen
network in bone, thus resulting in a decrease of bone toughness [71, 102–104].
4.1.1 Porosity
Loss of bone mass is a major age-related change in cortical bone, resulting mainly
from thinning of the cortex (reviewed in Whole-Bone Structure and Strength) and
increasing intracortical porosity [105–107]. For example, a study of human
cadaveric humeral cortex reported that mean cortical porosity increases with age,
70 X. Wang
from about 4% at 40 years of age to more than 10% at age 80, whereas the ‘‘true
mineral density’’, i.e., the density of the solid bone phase, does not vary signifi-
cantly with age [108]. Similar findings were reported for human femoral cortical
bone [109]. Thus, the main factor in age-related intracortical bone loss is the
increased porosity. Other two-dimensional morphological studies have indicated that
it is the increased pore area (i.e., Haversian canal size) rather than pore number that
explains most of the age-related increase in cortical porosity [158–160]. A recent
three-dimensional lCT study of human cadaveric femurs indicated that Haversian
canal volume fraction (porosity) increases with age (18–92 years); the number of
canals increased until the 6th decade then decreased [110].
4.1.2 Mineralization
Data on mineralization of cortical bone are mixed with respect to age and anatomic
site-dependence. Here we define mineralization as the percent of the solid phase of
bone that is mineral; a related measure is tissue mineral density (TMD) which is
mineral content per volume of solid tissue. These are not the same as ‘‘bone
mineral density (BMD)’’ which is defined as mineral content per total area
(aBMD) or volume (vBMD), and is thus influenced by porosity in bone as well as
tissue mineralization. Based on samples of femoral cortical bone from men, tissue
mineral density becomes less uniform with age, changing from predominantly low
density (*2.0 g/cc) in young adult (20–25 year old) to an increased fraction of
high density (2.2–2.3 g/cc) in elderly (80–85 years old) [111]. Age-dependent
increases in average mineralization of bone were also noted in cancellous bone
samples from women and men (18–96 years) [112]. Moreover, a backscattered
electron imaging study showed that the portion of hypermineralized tissue
increases significantly with age (in men and women), although the change was
greater in femoral neck and intertrochanteric cortices than in the femoral diaphysis
[105]. Consistent with the above studies, quantitative microradiography indicated
greater heterogeneity of mineralization with increasing age, although only in
women [100]. On the other hand, this same study reported that the average degree
of mineralization of femoral cortical bone decreased with age in women and did
not change in men [100]. Another study reported no age-related increase in cal-
cium concentration of human femoral bone (male and female donors). Taken
together, the data indicate no strong effect of age on the average mineralization of
cortical bone, although there appears to be greater heterogeneity with age.
Mineralization of bone is not uniform within the tissue and reflects a dynamic
process. Raman microscopy analysis of cortical bone samples from male donors
ranging between 17 and 73 years old demonstrated that the compartments of
primary lamellar bone (i.e., not replaced through remodeling process) grow older
through continuous maturation and growth of mineral crystals that may persist as
long as two decades [113]. Bone remodeling can remove such ‘‘elderly’’ tissues,
thus impeding the tissue aging process and maintaining the average mineralization
of tissue approximately constant over time. However, reduced bone remodeling
with age may lead to a net increase in tissue mineralization due to accumulation of
elderly tissue fragments [113].
Bone mineral also displays osteoporosis-related alterations in elemental com-
position and crystallinity. Biochemical analysis of iliac crest biopsies from oste-
oporotic subjects exhibited mineral with decreased CO3 and Ca/P ratio [107],
while Cohen and Kitzes found decreased Mg content [114]. Moreover, a Fourier
Transform Infrared Microscopy (FTIRM) study indicated that fracture risk was
increased with the mineral-to-matrix ratio and crystallinity [115]. For cortical bone
from iliac crest biopsies, the mineral phase was more crystalline/mature in samples
from osteoporotic donors than bone from normals [116]. (Changes in bone mineral
are also reviewed in Changes in Cortical Bone Mineral and Microstructure with
Aging and Osteoporosis).
4.1.3 Collagen Matrix
The age-related changes causing a functional deficiency of collagen are primarily

due to increased intermolecular cross-linking [99, 117–124]. The intermolecular
cross-linking of collagen molecules within the tissues involves two different
mechanisms: enzymatic and non-enzymatic crosslinks. Among the two, the non-
enzymatic crosslinking, known as glycation, is the major cause of dysfunction of
collagen matrix associated with aging. The non-enzymatic crosslinking involves
reaction with glucose and subsequent oxidation products of the complex. The
process may be accelerated in diabetic patients due to higher levels of serum
glucose. Non-enzymatic crosslinks are actually the outcome of so-called Advanced
Glycation End Products (AGEs). Pentosidine, one of collagen crosslinks by AGEs
found in bone, is commonly used as a biomarker of AGEs [125]. Higher urinary
pentosidine levels were associated with bone fractures in elderly diabetic patients
and may in part account for the reduced bone strength in type 2 diabetes [102].
In addition, aging may lead to a decrease in collagen content, thereby resulting in
the increased stiffness, enzyme resistance, and permeability and the decreased
swelling of bone [112]. Moreover, the effect of glycation on cell–matrix interac-
tions has also shown to be an equally important aspect of aging of bone [23].
Recent findings provide important evidence that bone proteins are affected by
AGEs modification, showing that glycation of bone matrix influences osteoclasts
(bone resorption) and osteoblasts (bone formation) [126]. Finally, recent studies
show that tissue age-dependent collagen maturity may also be correlated to frac-
ture risk of bone [115].
As the major part of extracellular matrix of bone, the thermostability of the
collagen network decreases with increasing age in terms of shrinking and melting
temperatures. Age related decreases in collagen shrinkage temperature have been
correlated to decreased bone toughness, thus substantiating the view that detri-
mental changes in the collagen network may lead to increased bone fragility with
aging [127].
72 X. Wang
Fig. 14 Age-related changes

in stiffness loss of human
cortical bone (diaphyseal
tibia) in tension. Stage I and
Stage II are the transition and
saturation region of bone
viscous responses,
respectively. ey indicates
the yield strain
4.2 Age-Related Changes in Mechanical Properties of Bone
4.2.1 Elastic Properties
Because of the well-documented increase in cortical porosity with age (reviewed

in the previous section), and the strong, negative relationship between elastic
modulus and porosity [56], it is expected that cortical bone elastic modulus
declines with age. Yet available data indicate either no change or, at most, a slight
decrease in the elastic modulus of human cortical bone with age. For example,
reports of age-related decreases in the elastic modulus of human femoral cortical
bone range between 1.5 and 2.3% per decade from a maximum value at *35 years
of age [60, 128], although one study reported no significant decline from age 20 to
102 [109]. Results for tibial bone are mixed; one study reported an age-related
increase in tensile modulus of *1.5% per decade, but a decrease in compressive
modulus of *4.0% per decade [60]. Note that this result [60] is questionable,
because it is not consistent with the well-supported conclusion by the same
investigators that the tensile and compressive moduli of cortical bone do not
differ [62].
Elastic modulus loss for cortical bone with respect to applied strain is not
significantly correlated with age. For example, a study on human cadaveric
femora from middle aged (50–60 years old) and elderly ([70 years old) donors
indicates that no significant difference exists in the curve of elastic modulus
versus applied strain except that the elderly (old aged) bone breaks at much
smaller strain compared with middle aged bone (Fig. 14). This also can be
verified by the elastic strain energy released by newly formed damage surfaces
as shown in Fig. 15, indicating that the trend of released elastic strain energy
dissipation with respect to the applied strain is similar for both age groups
(i.e., middle aged vs. elderly).

in the released elastic strain
energy of human cortical
bone (diaphyseal tibia) in
tension. No significant
changes exist except that
elderly bone breaks at a
significantly reduced applied
strain
4.2.2 Strength
There is clear evidence that human cortical bone strength deteriorates with age by
*2–5% per decade. For example, a study on bone obtained from donors ranging
from 20 to 98 years old reported that femoral yield strength (tension) decreases
2.2% per decade, while tibial yield strength decreases 0.5% per decade [60]. In the
same report, the ultimate tensile strength of femoral and tibial cortical bone
decrease at rates of 2.1 and 1.2% per decade, respectively, while the compressive
strengths decrease at 2.5 and 2.0% per decade, respectively [60]. In addition, a
study on cadaveric femurs ranging between 20 to 102 years old reported that
tensile ultimate stress and ultimate strain decrease by 5 and 9% per decade,
respectively [109]. These authors also reported that changes in porosity account
for *76% of the reduction in bone strength, whereas changes in calcium content
(a measure of mineralization) play a minor role in age-related changes. Similarly,
another study reported that the tensile strength of cortical bone decreases by 3.7%
per decade with increasing age [128]. In bending, cortical bone strength is
diminished by about 15–20% between the ages of 35 and 70, equivalent to
4.3–5.7% reduction per decade [129].
4.2.3 Toughness
Measures of cortical bone toughness decline with age at a faster rate than measures
of strength. For example, toughness (energy absorption to failure) was reported to
drop 12% per decade from 20 to 102 years old [109]. Similarly, it is reported that
the critical stress intensity factor (KC) falls by 4.1% per decade, J-integral by 3%
per decade, and the work to fracture (Wf) by 8.7% per decade [128]. Such dramatic
age-related changes in bone toughness are more determined by the final failure
strain rather than the mechanism of plastic energy dissipation as a function of
74 X. Wang
Fig. 16 Plastic energy

dissipation of bone as a
function of applied strain for
middle aged (45–55 years
old) and elderly bone
([70 years old)
applied strain [84]. A study on human cadaveric tibia in tension indicates that the
plastic energy dissipation is linearly related to the applied strain with a similar
slope for both age groups (i.e., middle aged and elderly) except that the elderly
bones break at much smaller strain levels (\2%) compared with the middle aged
ones (3–4%) (Fig. 16). Another study reported that more microdamage is formed
in elderly bone compared to its younger counterpart under the same loading
conditions [130].
Age-related changes in bone toughness may be related to changes in the
collagen matrix. A study on cortical bone of human cadaveric femora confirmed
age-dependent decreases in strength, work to fracture, and fracture toughness of
normal bone samples [25]. Importantly, elastic modulus, strength, and work to
fracture of the collagen network (demineralized bone samples) all decreased with
age, while the concentration of pentosidine (a marker of non-enzymatic glycation)
and bone porosity increased with age. Regression analyses indicate that the age-
related decrease in the toughness of bone (especially its post-yield portion) is
correlated significantly with deterioration of the mechanical integrity of the col-
lagen network, leading to the conclusion that the strength of the collagen network
decreases with age and correlates with bone toughness [25]. The results of a
Raman spectroscopy study on human cortical bone from donors over a wide age
range (34–99 years) supports the above conclusion, showing that possible changes
in collagen cross-linking chemistry depend on age, and correlate with the age-
related decrease in the toughness of bone [131].
5 Gender Differences in Human Cortical Bone
Women are more susceptible than men to age-related fractures. Bone fragility
depends on structural characteristics (e.g., bone size/shape) and tissue properties
(e.g., strength, fracture toughness). Available evidence indicates that the higher
fracture incidence in women most likely results mainly from quantitative differ-
ences in bone structure, rather than from difference in tissue properties or other
risk factors [132, 133]. Nonetheless, additional work is needed to improve our
understanding of gender differences in bone, to aid development of clinical and
pharmacological strategies to address the gender disparity.
5.1 Gender Difference in Cortical Bone Structural

and Composition
5.1.1 Bone Mass and Size
Peak cortical bone mass and areal BMD are higher in most men than women due to
bigger bones in men throughout life [134–136]. (This topic is reviewed in more detail
in Whole-Bone Structure and Strength and Factor of Risk for Fracture). Men
experience less age-related cortical bone loss due mostly to a slower rate of endosteal
resorption. For example, medullary expansion and thinning of the proximal femur
occurs faster in women than in men, thus leading to lower section modulus (related to
whole-bone strength) in women with increasing age [137]. Consistent with these
gender differences in structure, a cross-sectional dual-energy x-ray absorptiometry
(DEXA) study of 1,087 healthy adults (273 men and 814 women) aged 65–87 years
indicates that women experience age-related declines in BMD at all non-spine
skeletal sites, with the largest decline at the femoral neck (-0.0038 g/cm2/year) and
the smallest at the trochanter of the hip (-0.0023 g/cm2/year), whereas men do not
show significant changes at non-spine sites [138]. Even though there are gender
differences in bone geometry and thus areal BMD, the incidence of fracture seems to
be similar in men and women that have a similar absolute areal BMD [139].
Both men and women show evidence of periosteal expansion with age. For
example, a single-photon absorptiometry study of the distal radius of 108 women
followed over a mean period of 15 years after menopause indicates that the mean
(±SD) annual decrease in areal BMD is 1.9 ± 0.7%, while medullary bone
diameter increases 1.1 ± 0.9%, and periosteal diameter increases 0.7 ± 0.3% per
year. These results suggest that the increased bone loss after menopause is asso-
ciated with increased periosteal apposition, which may partially preserve whole-
bone strength in the setting of endosteal bone loss [140].
5.1.2 Porosity
The most significant age-related change in cortical bone microstructure is the

increased porosity, which affects both women and men. In some (but not all) studies,
bones from women tend to have larger Haversian canal diameter and greater overall
porosity relative to men [110]. Dramatic increases of porosity with aging have been
observed in various anatomical locations of the female skeleton, e.g. at the femoral
76 X. Wang
Fig. 17 3D lCT image of the femoral neck cortex in females: a the porosity is 8.66%. b the
porosity is 23.72% (Presented by Bousson et al. [141] in ASBMR)

in intracortical porosity of the
mid-shaft of human femur did
not differ between genders.
The means for subjects are
classified by decade. N
denotes the number of
subjects in each decade.
Feik et al. [142]
neck and intertrochanter [105]. Using a high resolution CT system with synchrotron
radiation, the porosity of female femoral neck cortex was observed to change
with age from about 5.0 to 39% of the total cortical bone volume (Fig. 17) [141].
In another study, the average porosity of femoral diaphyses increased from *5% in
young adults to over 9% in the elderly, although there were no significant gender
differences (Fig. 18) [142]. In addition, the porosity in humerus increased from 4% at
40 years of age to 10% and more at age 80 in both women and men [108].
5.1.3 Mineralization
There is some evidence of gender differences in bone mineralization with age,

although not all studies support a gender effect. An early study indicated that the
areal BMD of total body is greater for men than women, while average tissue
Fig. 19 Collagen orientation distribution in the cross-section of female femurs. a a 28-year-old

female individual demonstrating a high proportion of transverse collagen fibers, predominantly
located within circumferential lamellar bone, b a 51-year-old female individual demonstrating a
high proportion of transverse collagen fibers in circumferential lamellar bone of the anterior
cortex, c an 88-year-old female demonstrating little patterning of collagen fiber orientation
around the cortex. (Presented by Goldman et al. [143])
mineral density of the humeral shaft does not vary significantly between genders
[108]. Similarly, gender effects on mineralization were not significant in the
proximal and mid-shaft of the femur [105]. In slight contrast, quantitative
microradiography of the femoral shaft shows that women start with a higher level
of mineralization than men, but fall below the level in men after the sixth decade;
mineralization was more stable throughout life in men [100].
5.1.4 Collagen Phase
The effect of gender on the micro- and ultrastructural changes in the collagen network
of bone are not well documented in the literature. Nonetheless, some studies have
indicated that gender has an influence on collagen fibril orientation, collagen network
integrity, as well as its interaction with the mineral phase. The variability of preferred
collagen fiber orientation in bone from women is related to age, showing a complex
pattern between age groups (Fig. 19) [143]. Irrespective of a similar trend between
males and females, higher variability between different age groups is found in females
than males. For instance, the proportion of collagen fibers aligned transversely
decreases more with age in women. In addition, females show a higher proportion of
transversely oriented lamellae in newly formed bone than males [143, 144].
Cortical bone from women and men has shown age-related changes in both
immature and mature enzymatic collagen crosslinks. In a study on secondary osteons
and interstitial tissues from middle-aged (42–63 years) and elderly (69–90 years)
bone, the mature enzymatic cross-links (i.e., hydroxylysyl-pyridinoline [HP] and
lysylpyridinoline [LP]) decreased slightly after middle-age, perhaps due to increased
78 X. Wang
Table 5 Gender dependence of collagen crosslinks in human cortical bone [117]

HP LP PE
Female-middle [ Male-middle Female-middle [ Male-middle Female-middle \ Female-old
Female-old [ Male-old Female-old [ Male-old Male-middle \ Male-old
Female-middle [ Female-old
Male-middle [ Male-old
remodeling which results in less crosslinked collagen and/or due to decreased

enzymatic activity of lysyl oxidase with age [117]. In addition, the amount of non-
enzymatic crosslinking (pentosidine [PE]) increased from middle-age to elderly in
both women and men. Gender effects were small, with slightly increased HP and LP
in women (Table 5) [117]. Notably, differences between osteonal and interstitial
tissues were greater than those induced by age- or gender. Finally, the increased
variability of shrinkage temperature of the collagen network in bone may be gender-
dependent, decreasing markedly with age in men, but not women [145].
5.2 Gender Differences in Mechanical Properties

of Cortical Bone
There is limited information regarding the effect of gender on tissue properties of

human cortical bone. An early study reported no significant differences in the
mechanical properties of cortical bone between males and females, in which the
bone specimens from human femora and tibiae were tested in tension, torsion, and
compression for a population ranging in age from 20 to 86 years [60]. In addition,
a more recent study observed that bone matrix apparent stiffness was not signifi-
cantly different between males and females for femoral cortical bone [146].
Moreover, a fracture toughness study of human cortical bone indicates that tissue
toughness gradually decreases with age between 55 and 89 years old and no
significant differences in the toughness of bone exist between men and women
[147]. Thus, available evidence suggests that gender per se may not have signif-
icant effects on the tissue properties of cortical bone.
References
1. Martin, R.B., Ishida, J.: The relative effects of collagen fiber orientation, porosity, density,
and mineralization on bone strength. J. Biomech. 22(5), 419–426 (1989)
2. Jackson, S.A.: The fibrous structure of bone determined by x-ray diffraction. J. Biomed.
Eng. 1(2), 121–122 (1979)
3. Bembey, A.K., et al.: Contribution of collagen, mineral and water phases to the nanomechanical
properties of bone. In: Fundamentals of Nanoindentation and Nanotribology III, Symposium,
29 Nov.– 3 Dec. 2004, Materials Research Society, Boston, 2005
4. Buckwalter, J.A., Cooper, R.R.: Bone structure and function. Instr. Course Lect. 36, 27–48
(1987)
5. Frasca, P., Harper, R.A., Katz, J.L.: Scanning electron microscopy studies of collagen,
mineral and ground substance in human cortical bone. Scan. Electron Microsc. (Pt 3),
339–346 (1981)
6. Robinson, R.A.: Bone tissue: composition and function. Johns Hopkins Med. J. 145(1),
10–24 (1979)
7. Aerssens, J., et al.: Interspecies differences in bone composition, density, and quality:
potential implications for in vivo bone research. Endocrinology 139(2), 663–670 (1998)
8. Glimcher, M.J.: The nature of the mineral component of bone and the mechanism of
calcification. Instr. Course Lect. 36, 49–69 (1987)
9. Wachtel, E., Weiner, S.: Small-angle x-ray scattering study of dispersed crystals from bone
and tendon. J. Bone Miner. Res.: Off. J. Am. Soc. Bone Miner. Res. 9(10), 1651–1655 (1994)
10. Fratzl, P., et al.: Mineral crystals in calcified tissues: a comparative study by SAXS. J. Bone
Miner. Res. 7(3), 329–334 (1992)
11. Landis, W.J., Glimcher, M.J.: Electron diffraction and electron probe microanalysis of the
mineral phase of bone tissue prepared by anhydrous techniques. J. Ultrastruct. Res.
63, 188–223 (1978)
12. Roseberry, H., Hastings, A., Morse, J.: X-ray analysis of bone and teeth. J. Biol. Chem.
90, 395–407 (1931)
13. Posner, A.S.: Crystal chemistry of bone mineral. Physiol. Rev. 49(4), 760–792 (1969)
14. Boskey, A.: Bone mineral crystal size. Osteoporos. Int. 14, S16–S21 (2003)
15. Rey, C., et al.: Hydroxyl groups in bone mineral. Bone 16(5), 583–586 (1995)
16. Loong, C.K., et al.: Evidence of hydroxyl-ion deficiency in bone apatites: an inelastic
neutron-scattering study. Bone 26(6), 599–602 (2000)
17. Nimni, B.S., et al.: Changes in the ratio of non-calcified collagen to calcified collagen in
human vertebrae with advancing age. Connect. Tissue Res. 29(2), 133–140 (1993)
18. Tzaphlidou, M.: Bone architecture: collagen structure and calcium/phosphorus maps.
J. Biol. Phys. 34(1–2), 39–49 (2008)
19. Yamauchi, M.: The major matrix molecule in mineralized tissues. In: Anderson, J.J.,
Garner, S.C. (eds.) Calcium and Phosphorus in Health and Disease, pp. 127–145. CRC
Press, Boca Raton (1996)
20. Bailey, A.J., Knott, L.: Molecular changes in bone collagen in osteoporosis and
osteoarthritis in the elderly. Exp. Gerontol. 34(3), 337–351 (1999)
21. Eyre, D.R., Dickson, I.R., Van Ness, K.: Collagen cross-linking in human bone and articular
cartilage. Age-related changes in the content of mature hydroxypyridinium residues.
Biochem. J. 252(2), 495–500 (1988)
22. Bunn, H.F., Gabbay, K.H., Gallop, P.M.: The glycosylation of hemoglobin: relevance to
diabetes mellitus. Science 200(4337), 21–27 (1978). (New York)
23. Bailey, A.J., Paul, R.G., Knott, L.: Mechanisms of maturation and ageing of collagen.
Mech. Ageing Dev. 106(1–2), 1–56 (1998)
24. Vashishth, D.: Collagen glycation and its role in fracture properties of bone.
J. Musculoskelet. Neuronal Interact. 5(4), 316 (2005)
25. Wang, X.: Age-related changes in the collagen network and toughness of bone. Bone 31(1),
1–7 (2002)
26. Ingram, R.T., et al.: Distribution of noncollagenous proteins in the matrix of adult human
bone: evidence of anatomic and functional heterogeneity. J. Bone Miner. Res. 8(9),
1019–1029 (1993)
27. Mbuyi-Muamba, J.M., Dequeker, J., Gevers, G.: Collagen and non-collagenous proteins in
different mineralization stages of human femur. Acta Anat. (Basel) 134(4), 265–268 (1989)
28. Cowles, E.A., et al.: Mineralization and the expression of matrix proteins during in vivo
bone development. Calcif. Tissue Int. 62(1), 74–82 (1998)
29. Fisher, L.W., Termine, J.D.: Noncollagenous proteins influencing the local mechanisms of
calcification. Clin. Orthop. Relat. Res. 200, 362–385 (1985)
80 X. Wang
30. Roy, M.E., et al.: Correlations between osteocalcin content, degree of mineralization, and
mechanical properties of C. carpio rib bone. J. Biomed. Mater. Res. 54(4), 547–553 (2001)
31. Kavukcuoglu, N.B., Patterson-Buckendahl, P., Mann, A.B.: Effect of osteocalcin deficiency
on the nanomechanics and chemistry of mouse bones. J. Mech. Behav. Biomed. Mater. 2(4),
348–354 (2009)
32. Nyman, J.S., et al.: Measurements of mobile and bound water by nuclear magnetic
resonance correlate with mechanical properties of bone. Bone 42(1), 193–199 (2008)
33. Wilson, E.E., et al.: Three structural roles for water in bone observed by solid-state NMR.
Biophys. J. 90(10), 3722–3731 (2006)
34. Wilson, E.E., et al.: Highly ordered interstitial water observed in bone by nuclear magnetic
resonance. J. Bone Miner. Res.: Off. J. Am. Soc. Bone Miner. Res. 20(4), 625–634 (2005)
35. Nomura, S., et al.: Interaction of water with native collagen. Biopolymers 16(2), 231–246
(1977)
36. Pineri, M.H., Escoubes, M., Roche, G.: Water–collagen interactions: calorimetric and
mechanical experiments. Biopolymers 17(12), 2799–2815 (1978)
37. Giraud-Guille, M.M.: Twisted plywood architecture of collagen fibrils in human compact
bone osteons. Calcif. Tissue Int. 42(3), 167–180 (1988)
38. Weiner, S., et al.: Rotated plywood structure of primary lamellar bone in the rat:
orientations of the collagen fibril arrays. Bone 20(6), 509–514 (1997)
39. Weiner, S., Traub, W., Wagner, H.D.: Lamellar bone: structure-function relations. J. Struct.
Biol. 126(3), 241–255 (1999)
40. Weiner, S., Traub, W.: Organization of hydroxyapatite crystals within collagen fibrils.
FEBS Lett. 206(2), 262–266 (1986)
41. Katz, E.P., et al.: The structure of mineralized collagen fibrils. Connect. Tissue Res.
21(1–4), 149–154 (1989)
42. Sasaki, N., et al.: Atomic force microscopic studies on the structure of bovine femoral
cortical bone at the collagen fibril-mineral level. J. Mater. Sci. Mater. Med. 13(3), 333–337
(2002)
43. Lees, S.: Considerations regarding the structure of the mammalian mineralized osteoid from
viewpoint of the generalized packing model. Connect. Tissue Res. 16(4), 281–303 (1987)
44. Lee, D.D., Glimcher, M.J.: The three-dimensional spatial relationship between the collagen
fibrils and the inorganic calcium-phosphate crystals of pickerel and herring fish bone.
Connect. Tissue Res. 21(1–4), 247–257 (1989)
45. Balooch, M., et al.: Mechanical properties of mineralized collagen fibrils as influenced by
demineralization. J. Struct. Biol. 162(3), 404–410 (2008)
46. Jèager, I., Fratzl, P.: Mineralized collagen fibrils: a mechanical model with a staggered
arrangement of mineral particles. Biophys. J. 79(4), 1737–1746 (2000)
47. Pidaparti, R.M., et al.: Bone mineral lies mainly outside collagen fibrils: predictions of a
composite model for osteonal bone. J. Biomech. 29(7), 909–916 (1996)
48. Nikolov, S., Raabe, D.: Hierarchical modeling of the elastic properties of bone at submicron
scales: the role of extrafibrillar mineralization. Biophys. J. 94(11), 4220–4232 (2008)
49. Fantner, G.E., et al.: Sacrificial bonds and hidden length dissipate energy as mineralized
fibrils separate during bone fracture. Nat. Mater. 4(8), 612–616 (2005)
50. Wang, X., Nyman, J.S.: A novel approach to assess post-yield energy dissipation of bone in
tension. J. Biomech. 40(3), 674–677 (2007)
51. Jepsen, K.J., Davy, D.T.: Comparison of damage accumulation measures in human cortical
bone. J. Biomech. 30(9), 891–894 (1997)
52. Joo, W., Jepsen, K.J., Davy, D.T.: The effect of recovery time and test conditions on
viscoelastic measures of tensile damage in cortical bone. J. Biomech. 40(12), 2731–2737
(2007)
53. Espinoza Orías, A.A., et al.: Anatomic variation in the elastic anisotropy of cortical bone
tissue in the human femur. J. Mech. Behav. Biomed. Mater. 2(3), 255–263 (2009)
54. Reilly, D.T., Burstein, A.H.: The elastic and ultimate properties of compact bone tissue.
J. Biomech. 8(6), 393–405 (1975)
55. Dempster, W.T., Liddicoat, R.T.: Compact bone as a non-isotropic material. Am. J. Anat.
91(3), 331–362 (1952)
56. Schaffler, M.B., Burr, D.B.: Stiffness of compact bone: effects of porosity and density.
J. Biomech. 21(1), 13–16 (1988)
57. Wachter, N.J., et al.: Correlation of bone mineral density with strength and microstructural
parameters of cortical bone in vitro. Bone 31(1), 90–95 (2002)
58. Snyder, S.M., Schneider, E.: Estimation of mechanical properties of cortical bone by
computed tomography. J. Orthop. Res.: Off. Publ. Orthop. Res. Soc. 9(3), 422–431 (1991)
59. Ko, R.: The tension test upon the compact substance of the long bones of human
extremities. J. Kyoto. Pref. Med. Univ. 53(4), 503–525 (1953)
60. Burstein, A., Reilly, D., Martens, M.: Aging of bone tissue: mechanical properties. J. Bone
Jt. Surg. Am. 58(1), 82–86 (1976)
61. Hoffler, C.E., et al.: Heterogeneity of bone lamellar-level elastic moduli. Bone 26(6),
603–609 (2000)
62. Reilly, D.T., Burstein, A.H., Frankel, V.H.: The elastic modulus for bone. J. Biomech. 7(3),
271–275 (1974)
63. Sedlin, E.D.: A rheologic model for cortical bone. A study of the physical properties of
human femoral samples. Acta Orthop. Scand. Suppl. 83, 1–77 (1965)
64. Sedlin, E.D., Hirsch, C.: Factors affecting the determination of the physical properties of
femoral cortical bone. Acta Orthop. Scand. 37(1), 29–48 (1966)
65. Wang, X., et al.: The role of collagen in determining bone mechanical properties. J. Orthop.
Res. 19(6), 1021–1026 (2001)
66. Nyman, J.S.: The influence of water removal on the strength and toughness of cortical bone.
J. Biomech. 39(5), 931–938 (2006)
67. Ascenzi, A., Bonucci, E.: The tensile properties of single osteons. Anat. Rec. 158(4),
375–386 (1967)
68. Ascenzi, A., Bonucci, E.: The compressive properties of single osteons. Anat. Rec. 161(3),
377–391 (1968)
69. Ascenzi, A., Baschieri, P., Benvenuti, A.: The torsional properties of single selected
osteons. J. Biomech. 27(7), 875–884 (1994)
70. Edward Hoffler, C., et al.: An application of nanoindentation technique to measure bone
tissue lamellae properties. J. Biomech. Eng. 127(7), 1046–1053 (2005)
71. Nyman, J.S., et al.: Age-related factors affecting the postyield energy dissipation of human
cortical bone. J. Orthop. Res.: Off. Publ. Orthop. Res. Soc. 25(5), 646–655 (2007)
72. Pidaparti, R.M., Wang, Q.Y., Burr, D.B.: Modeling fatigue damage evolution in bone.
Biomed. Mater. Eng. 11(2), 69–78 (2001)
73. Reilly, G.C., Currey, J.D.: The effects of damage and microcracking on the impact strength
of bone. J. Biomech. 33(3), 337–343 (2000)
74. Fondrk, M.T., Bahniuk, E.H., Davy, D.T.: A damage model for nonlinear tensile behavior of
cortical bone. J. Biomech. Eng. 121(5), 533–541 (1999)
75. Burr, D.B., et al.: Does microdamage accumulation affect the mechanical properties of
bone? J. Biomech. 31(4), 337–345 (1998)
76. Akkus, O., et al.: Relationship between damage accumulation and mechanical property
degradation in cortical bone: microcrack orientation is important. J. Biomed. Mater. Res.:
Part A 65(4), 482–488 (2003)
77. Ebacher, V., et al.: Strain redistribution and cracking behavior of human bone during
bending. Bone 40(5), 1265–1275 (2007)
78. Diab, T., Vashishth, D.: Morphology, localization and accumulation of in vivo
microdamage in human cortical bone. Bone 40(3), 612–618 (2007)
79. Currey, J.D.: Tensile yield in compact bone is determined by strain, post-yield behaviour by
mineral content. J. Biomech. 37(4), 549–556 (2004)
80. Biewener, A.A.: Safety factors in bone strength. Calcif. Tissue Int. 53(1), S68–S74 (1993)
81. Niebur, G.L., et al.: High-resolution finite element models with tissue strength asymmetry
accurately predict failure of trabecular bone. J. Biomech. 33(12), 1575–1583 (2000)
82 X. Wang
82. Leng, H., Dong, X.N., Wang, X.: Progressive post-yield behavior of human cortical bone in
compression for middle-aged and elderly groups. J. Biomech. 42(4), 491–497 (2009)
83. Nyman, J.S., et al.: Differences in the mechanical behavior of cortical bone between
compression and tension when subjected to progressive loading. J. Mech. Behav. Biomed.
Mater. 2(6), 613–619 (2009)
84. Nyman, J.S., et al.: Mechanical behavior of human cortical bone in cycles of advancing
tensile strain for two age groups. J. Biomed. Mater. Res. A 89(2), 521–529 (2009)
85. Kotha, S.P., Guzelsu, N.: Tensile damage and its effects on cortical bone. J. Biomech.
36(11), 1683–1689 (2003)
86. Zioupos, P.: Ageing human bone: factors affecting its biomechanical properties and the role
of collagen. J. Biomater. Appl. 15(3), 187–229 (2001)
87. Vashishth, D., Tanner, K.E., Bonfield, W.: Experimental validation of a microcracking-
based toughening mechanism for cortical bone. J. Biomech. 36(1), 121–124 (2003)
88. Vashishth, D., Tanner, K.E., Bonfield, W.: Contribution, development and morphology of
microcracking in cortical bone during crack propagation. J. Biomech. 33(9), 1169–1174 (2000)
89. Parsamian, G.P., Norman, T.L.: Diffuse damage accumulation in the fracture process zone
of human cortical bone specimens and its influence on fracture toughness. J. Mater. Sci.
Mater. Med. 12(9), 779–783 (2001)
90. Yeni, Y.N., Norman, T.L.: Fracture toughness of human femoral neck: effect of microstructure,
composition, and age [published erratum appears in Bone 2000 Aug;27(2):327]. Bone 26(5),
499–504 (2000)
91. Kruzic, J.J., et al.: Crack blunting, crack bridging and resistance-curve fracture mechanics in
dentin: effect of hydration. Biomaterials 24(28), 5209–5221 (2003)
92. Yamashita, J., et al.: Collagen and bone viscoelasticity: a dynamic mechanical analysis.
J. Biomed. Mater. Res. 63(1), 31–36 (2002)
93. Johnson, M., Katz, J.L.: Some new developments in the rheology of bone. Biorheol. Suppl.:
Off. J. Int. Soc. Biorheol. 1, 169–174 (1984)
94. Garner, E., et al.: Viscoelastic dissipation in compact bone: implications for stress-induced
fluid flow in bone. J. Biomech. Eng. 122(2), 166–172 (2000)
95. Yeni, Y.N., et al.: Apparent viscoelastic anisotropy as measured from nondestructive
oscillatory tests can reflect the presence of a flaw in cortical bone. J. Biomed. Mater. Res:
Part A 69(1), 124–130 (2004)
96. Yeni, Y.N., et al.: The effect of yield damage on the viscoelastic properties of cortical bone
tissue as measured by dynamic mechanical analysis. J. Biomed. Mater. Res.: Part A 82(3),
530–537 (2007)
97. Sasaki, N., Enyo, A.: Viscoelastic properties of bone as a function of water content.
J. Biomech. 28(7), 809–815 (1995)
98. Sasaki, N., et al.: Stress relaxation function of bone and bone collagen. J. Biomech. 26(12),
1369–1376 (1993)
99. Avery, N.C., Bailey, A.J.: Enzymic and non-enzymic cross-linking mechanisms in relation
to turnover of collagen: relevance to aging and exercise. Scand. J. Med. Sci. Sports 15(4),
231–240 (2005)
100. Bergot, C., et al.: The degree and distribution of cortical bone mineralization in the human
femoral shaft change with age and sex in a microradiographic study. Bone 45(3), 435–442
(2009)
101. Seeman, E.: Bone quality: the material and structural basis of bone strength. J. Bone Miner.
Metab. 26(1), 1–8 (2008)
102. Schwartz, A.V., et al.: Pentosidine and increased fracture risk in older adults with type 2
diabetes. J. Clin. Endocrinol. Metab. 94(7), 2380–2386 (2009)
103. Saito, M., et al.: Role of collagen enzymatic and glycation induced cross-links
as a determinant of bone quality in spontaneously diabetic WBN/Kob rats. Osteoporos.
Int. 17(10), 1514–1523 (2006). Journal established as result of cooperation between
the European Foundation for Osteoporosis and the National Osteoporosis Foundation of
the USA
104. Wang, X., et al.: Age-related changes in the collagen network and toughness of bone. Bone
31(1), 1–7 (2002)
105. Vajda, E.G., Bloebaum, R.D.: Age-related hypermineralization in the female proximal
human femur. Anat. Rec. 255(2), 202–211 (1999)
106. Russo, C.R., et al.: Structural adaptations to bone loss in aging men and women. Bone
38(1), 112–118 (2006)
107. Burnell, J.M., et al.: Bone matrix and mineral abnormalities in postmenopausal
osteoporosis. Metabolism 31(11), 1113–1120 (1982)
108. Laval-Jeantet, A.M., et al.: Cortical bone senescence and mineral bone density of the
humerus. Calcif. Tissue Int. 35(3), 268–272 (1983)
109. McCalden, R.W., et al.: Age-related changes in the tensile properties of cortical bone. The
relative importance of changes in porosity, mineralization, and microstructure. J. Bone Joint
Surg. Am. 75(8), 1193–1205 (1993)
110. Cooper, D.M.L., et al.: Age-dependent change in the 3D structure of cortical porosity at the
human femoral midshaft. Bone 40(4), 957–965 (2007)
111. Simmons Jr, E.D., Pritzker, K.P., Grynpas, M.D.: Age-related changes in the human femoral
cortex. J. Orthop. Res.: Off. Publ. Orthop. Res. Soc. 9(2), 155–167 (1991)
112. Bailey, A.J., et al.: Age-related changes in the biochemical properties of human cancellous
bone collagen: relationship to bone strength. Calcif. Tissue Int. 65(3), 203–210 (1999)
113. Akkus, O., et al.: Aging of microstructural compartments in human compact bone. J. Bone
Miner. Res. 18(6), 1012–1019 (2003)
114. Cohen, L., Kitzes, R.: Infrared spectroscopy and magnesium content of bone mineral in
osteoporotic women. Isr. J. Med. Sci. 17(12), 1123–1125 (1981)
115. Gourion-Arsiquaud, S., et al.: Use of FTIR spectroscopic imaging to identify parameters
associated with fragility fracture. J. Bone Miner. Res.: Off. J. Am. Soc. Bone Miner. Res.
24(9), 1565–1571 (2009)
116. Paschalis, E.P., et al.: FTIR microspectroscopic analysis of human iliac crest biopsies from
untreated osteoporotic bone. Calcif. Tissue Int. 61(6), 487–492 (1997)
117. Nyman, J.S., et al.: Age-related effect on the concentration of collagen crosslinks in human
osteonal and interstitial bone tissue. Bone 39(6), 1210–1217 (2006)
118. Paschalis, E.P., et al.: Bone fragility and collagen cross-links. J. Bone Miner. Res.: Off.
J. Am. Soc. Bone Miner. Res. 19(12), 2000–2004 (2004)
119. Odetti, P., et al.: Role of advanced glycation end products in aging collagen. A scanning
force microscope study. Gerontology 44(4), 187–191 (1998)
120. Knott, L., Bailey, A.J.: Collagen cross-links in mineralizing tissues: a review of their
chemistry, function, and clinical relevance. Bone 22(3), 181–187 (1998)
121. Oxlund, H., Mosekilde, L., Ortoft, G.: Reduced concentration of collagen reducible cross
links in human trabecular bone with respect to age and osteoporosis. Bone 19(5), 479–484
(1996)
122. Danielsen, C.C., Mosekilde, L., Bollerslev, J.: Thermal stability of cortical bone collagen in
relation to age in normal individuals and in individuals with osteopetrosis. Bone 15(1),
91–96 (1994)
123. Walters, C., Eyre, D.R.: Collagen crosslinks in human dentin: increasing content of
hydroxypyridinium residues with age. Calcif. Tissue Int. 35(4–5), 401–405 (1983)
124. Nielsen, C.J., Bentley, J.P., Marshall, F.J.: Age-related changes in reducible crosslinks of
human dental pulp collagen. Arch. Oral Biol. 28(8), 759–764 (1983)
125. Odetti, P., et al.: Advanced glycation end products and bone loss during aging. Ann. N. Y.
Acad. Sci. 1043, 710–717 (2005)
126. Hein, G.E.: Glycation endproducts in osteoporosis–is there a pathophysiologic importance?
Clin. Chim. Acta 371(1–2), 32–36 (2006)
127. Zioupos, P.: The role of collagen in the declining mechanical properties of aging human
cortical bone. J. Biomed. Mater. Res. 45(2), 108–116 (1999)
128. Zioupos, P., Currey, J.D.: Changes in the stiffness, strength, and toughness of human
cortical bone with age. Bone 22(1), 57–66 (1998)
84 X. Wang
129. Martin, B.: Aging and strength of bone as a structural material. Calcif. Tissue Int. 53(1),
S34–S39 (1993). discussion S39–S40
130. Courtney, A.C., Hayes, W.C., Gibson, L.J.: Age-related differences in post-yield damage in
human cortical bone. Experiment and model. J. Biomech. 29(11), 1463–1471 (1996)
131. Ager, J.W., et al.: Deep-ultraviolet Raman spectroscopy study of the effect of aging on
human cortical bone. J. Biomed. Opt. 10(3), 034012 (2005)
132. Sigurdsson, G., et al.: Increasing sex difference in bone strength in old age: the age,
gene/environment susceptibility-reykjavik study (AGES-REYKJAVIK). Bone 39(3),
644–651 (2006)
133. Duan, Y., Seeman, E., Turner, C.H.: The biomechanical basis of vertebral body fragility in
men and women. J. Bone Miner. Res.: Off. J. Am. Soc. Bone Miner. Res. 16(12), 2276–2283
(2001)
134. Looker, A.C., et al.: Age, gender, and race/ethnic differences in total body and subregional
bone density. Osteoporos. Int. 20(7), 1141–1149 (2009)
135. Lei, S.F., et al.: Bone mineral density in elderly Chinese: effects of age, sex, weight, height,
and body mass index. J. Bone Miner. Metab. 22(1), 71–78 (2004)
136. Yan, L., et al.: Age- and gender-related differences in bone mineral status and biochemical
markers of bone metabolism in Northern Chinese men and women. Bone 30(2), 412–415
(2002)
137. Kaptoge, S., et al.: Effects of gender, anthropometric variables, and aging on the evolution
of hip strength in men and women aged over 65. Bone 32(5), 561–570 (2003)
138. Krall, E.A., et al.: Bone mineral density and biochemical markers of bone turnover in
healthy elderly men and women. J. Gerontol. Ser. A: Biol. Sci. Med. Sci. 52(2), M61–M67
(1997)
139. Kaufman, J.M., et al.: Background for studies on the treatment of male osteoporosis: state of
the art. Ann. Rheum. Dis. 59(10), 765–772 (2000)
140. Ahlborg, H.G., et al.: Bone loss and bone size after menopause. N. Engl. J. Med. 349(4),
327–334 (2003)
141. Bousson, V., et al.: Cortical bone in the human femoral neck: three-dimensional appearance
and porosity using synchrotron radiation. J. Bone Miner. Res. 19(5), 794–801 (2004)
142. Feik, S.A., Thomas, C.D., Clement, J.G.: Age-related changes in cortical porosity of the
midshaft of the human femur. J. Anat. 191(Pt 3), 407–416 (1997)
143. Goldman, H.M., et al.: Preferred collagen fiber orientation in the human mid-shaft femur.
Anat. Rec. Part A: Discov. Mol. Cell. Evol. Biol. 272(1), 434–445 (2003)
144. Goldman, H.M., et al.: Relationships among microstructural properties of bone at the human
midshaft femur. J. Anat. 206(2), 127–139 (2005)
145. Danielsen, C.C.: Thermal stability of cortical bone collagen in relation to age in normal
individuals and in individuals with osteopetrosis. Bone 15(1), 91–96 (1994)
146. Yeni, Y.N., Vashishth, D., Fyhrie, D.P.: Estimation of bone matrix apparent stiffness variation
caused by osteocyte lacunar size and density. J. Biomech. Eng. 123(1), 10–17 (2001)
147. Norman, T.L., Nivargikar, S.V., Burr, D.B.: Resistance to crack growth in human cortical
bone is greater in shear than in tension. J. Biomech. 29(8), 1023–1031 (1996)
148. Vincentelli, R., Grigorov, M.: The effect of Haversian remodeling on the tensile properties
of human cortical bone. J. Biomech. 18(3), 201–207 (1985)
149. McElhaney, J.H.: Dynamic response of bone and muscle tissue. J. Appl. Physiol. 21(4),
1231–1236 (1966)
150. Keller, T.S., Mao, Z., Spengler, D.M.: Young’s modulus, bending strength, and tissue
physical properties of human compact bone. J. Orthop. Res.: Off. Publ. Orthop. Res. Soc.
8(4), 592–603 (1990)
151. Smith, J.W., Walmsley, R.: Factors affecting the elasticity of bone. J. Anat. 93, 503–523 (1959)
152. Weiner, S., Wagner, H.D.: The material bone: structure-mechanical function relations.
Annu. Rev. Mater. Sci. 28, 271–298 (1998)
153. Ascenzi, A., Baschieri, P., Benvenuti, A.: The bending properties of single osteons.
J. Biomech. 23(8), 763–771 (1990)
154. Choi, K., et al.: The elastic moduli of human subchondral, trabecular, and cortical bone tissue
and the size-dependency of cortical bone modulus. J. Biomech. 23(11), 1103–1113 (1990)
155. Choi, K., Goldstein, S.A.: A comparison of the fatigue behavior of human trabecular and
cortical bone tissue. J. Biomech. 25(12), 1371–1381 (1992)
156. Jepsen, K.J., et al.: Type I collagen mutation alters the strength and fatigue behavior
of Mov13 cortical tissue. J. Biomech. 30(11–12), 1141–1147 (1997)
157. Wang, X., et al.: The role of collagen in determining bone mechanical properties. J. Orthop.
Res. 19(6), 1021–1026 (2001)
158. Brockstedt, H., et al.: Age-related and sex-related changes in iliac cortical bone mass and
remodeling. Bone 14(4), 681–691 (1993)
159 Stein, M.S., et al.: An automated analysis of intracortical porosity in human femoral bone
across age. J Bone Miner Res. 14(4), 624–632 (1999)
160. Thomas, C.D.L., Feik, S.A., Clement, J.G.: Increase in pore area, and not pore density, is the
main determinant in the development of porosity in human cortical bone. J. Anat. 209(2),
219–230 (2006)
Bone Microdamage and Its Contributions
to Fracture
Lamya Karim and Deepak Vashishth
Abstract Microdamage formation is a major determinant of bone fracture. The

nature and type of damage formed, as linear microcracks or diffuse damage,
depends on the interaction between applied loading and the extracellular matrix.
Human bone naturally experiences multi-axial cyclic loading. Changes in its
extracellular matrix can contribute to the overall deterioration of bone’s
mechanical integrity with aging and/or disease. This chapter provides a review of
literature reports on the detection of microdamage and its limitations; alterations in
microdamage with aging and disease; differences in microdamage between gender
and bone’s two distinct structural forms (cancellous and cortical); and the role of
microdamage in bone’s mechanical properties.
1 Introduction
Microdamage forms in composite materials, such as bone, due to loading and is an

indicator of the fracture process and eventual failure of the material. In vivo cyclic
loads lead to incremental formation of microdamage through fatigue. The fatigue
process results in small cracks that accumulate in the mineralized matrix of bone,
which can alter bone’s mechanical properties. Hence, the increased number of
cracks combined with further load bearing can result in failure [1, 2]. Although
microdamage occurrence is common in many load-bearing materials that are
subject to repetitive stresses, bone has the ability to remove microdamage through
L. Karim D. Vashishth (&)

Department of Biomedical Engineering,
Center for Biotechnology and Interdisciplinary Studies,
Rensselaer Polytechnic Institute, Troy, NY, USA
e-mail: vashid@rpi.edu

DOI: 10.1007/8415_2011_107
Published Online: 20 October 2011
88 L. Karim and D. Vashishth
Fig. 1 Microdamage forms as two different morphologies. An image of a linear microcrack (top)
is shown under a bright-field microscopy, and b confocal microscopy. An image of diffuse
damage (bottom) is shown under c bright-field microscopy, and d confocal microscopy. Scale
bars = 50 lm. Reprinted with permission from Elsevier [8]
a repair process of intracortical remodeling [3–5]. However, there is an age-related

accumulation of microdamage in bones [1, 6], due either to deterioration of the
repair mechanism with age [4, 7] or to matrix changes that reduce the damage
resistance of bone.
Microdamage can take two distinct forms [1, 6, 8]. Two damage morphologies,
linear microcracks and diffuse damage (Fig. 1), result from different types of
applied loading [9–11]. Linear microcracks form primarily due to compressive
loading and appear as sharply defined cracks [10, 12, 13]. They are primarily
found in the interstitial regions of bone where they follow the lamellar interface
and stop at the cement lines of osteons [8, 11, 14]. On the other hand, diffuse
damage results from tensile loads [9]. It has the appearance of a spread mesh of
submicroscopic cracks [15]. Diffuse damage is closely associated with osteonal
regions in bone, and it does not follow the lamellar boundaries [8].
Bone Microdamage and its Contributions to Fracture 89
Both linear microcracks and diffuse damage are typically identified using
histology methods. Along with identification of microdamage types, the rela-
tionship between microdamage and aging has been established, and the role of
microdamage in fracture toughness has also been investigated. This chapter will
review the literature on microdamage detection, changes in microdamage with
aging and/or disease or due to changes in bone quality, differences found in
microdamage between genders and between cancellous and cortical bone types,
and the role of microdamage in bone fragility.
2 Detection of Microdamage
Histology methods are typically used to identify in vivo microdamage. One

commonly used technique was originally developed by Frost [16] and later
modified by Burr and Stafford [12]. This technique involves en bloc staining of
bone tissue sections in a 1% basic fuchsin solution based in increasing concen-
trations of ethanol (70, 80, 90 and 100%) in vacuum at room temperature for a
period of five days. The en bloc stained cross sections are then embedded in
poly(methyl methacrylate), serially sectioned to 200 lm thickness, and ground to
100 lm thickness. Thus, only microdamage present in bone at the time of donor
death (i.e. before sectioning) is marked with basic fuchsin. The stained sections
can then be analyzed under a transmitted light microscope to determine the nature
of induced microdamage. Numerous groups have used the above staining method
with traditional fluorescence microscopy or laser confocal microscopy to observe
both linear microcracks and diffuse damage [1, 6, 15–18]. Under bright transmitted
light, linear microcracks appear as sharply defined lines with edges that are more
intensely stained than the surrounding space [10, 12, 19]. In contrast, diffuse
damage appears as a focal yet diffused area of pooled staining [15].
Another technique for marking microdamage incorporates staining with a single
chelating agent (e.g. alizarin complexone, calcein, calcein blue, xylenol orange,
oxytetracylcine). This method involves immersion of the specimen in a 1% stain
solution for a period of 20 h with a solution change after the first 4 h. Specimens
are then rinsed in distilled water to remove excess stain before sectioning and
analysis via microscopy. This procedure was found to be equally effective as the
basic fuchsin technique in identification of in vivo microdamage although the
labeling mechanisms between the two approaches differ [19, 20]. Basic fuchsin
fills in void spaces [12, 16, 21] while chelating agents bind to free floating calcium
ions within the damaged areas [22]. Although any of these methods can be used to
identify native microdamage, multi-labeling with several dyes allows for differ-
entiation between microdamage produced at different time points of the failure
process. One example of a multi-labeling procedure involves staining in 1%
oxytetracycline for 16 h under vacuum to first stain in vivo microdamage. Spec-
imens are subsequently immersed in calcein blue after the first 75% of the fatigue
test and then immersed in xylenol orange after the last 25% of the test (Fig. 2).
Fig. 2 A microcrack (light

blue and white) was labeled
with calcein blue after the
first 75% of fatigue testing
followed by xylenol orange
after the last 25% of the test.
Reprinted with permission
from John Wiley & Sons [20]
Following the completion of the fatigue testing, specimens are incubated in

ethanol for dehydration, embedded in methyl methacrylate, and ground to a
thickness of 100 lm. Results obtained during the development of this protocol
showed microcrack growth during the duration of fatigue testing [20], and
illustrated the use of dye combinations to capture microdamage development and
accumulation [23].
The imaging techniques for microdamage characterization discussed above
are limited to two-dimensional histological sectioning although microdamage
extends in three dimensions. Recently, three-dimensional methods for microdamage
evaluation have been described. One method is based on a novel automated fluo-
rescent imaging method that was recently developed by Kazakia et al. [24]. Bone
specimens are stained using fluorescent dyes and are embedded in a resin. Using a
computer numeric controlled (CNC) mill, the specimen surface is serially sectioned
in small increments. Fluorochromes that are exposed after the milling of each
section are excited and hence, fluorescence imaging allows for image aquisition of
the evenly spaced two-dimensional sections. The two-dimensional images are
reconstructed into a three-dimensional model in which stained components can be
visualized. This technique is an improvement over available histological micro-
damage quantification techniques. However, it is currently limited to overall
microdamage quantification and cannot be used to distinguish between morpholo-
gies of microdamage [25]. We have described a second method, a non-invasive
microcomputed tomography based technique using a heavy metal stain to charac-
terize microdamage quantity and morphology [26–28]. The staining procedure,
modified from a previous protocol [29], involves a 14 day immersion of specimens
in an equal mixture of 8% uranyl acetate in 70% acetone and 20% lead acetate in
70% acetone, followed by a 7 day immersion in 1% ammonium sulfide in acetone
[27, 28]. The central cubic region of stained specimens is scanned by microcom-
puted tomography. Scanning electron microscopy studies have confirmed that
regions of heavy metal staining correspond to areas of microdamage in bone (Fig. 3)
[27]. From the regions stained with heavy metal, the ratios of damaged volume to
bone volume (DV/BV) and damaged surface area to damaged bone volume (DS/
DV) can be calculated. Here, an increase in DV/BV represents an increase in mi-
crodamage quantity. The surface-to-volume ratio of the microdamage, DS/DV,
illustrates microdamage morphology where higher DS/DV represents linear mi-
crocracks and lower DS/DV represents diffuse damage (Fig. 4) [26, 28]. A DS/DV
based numerical index may help to eliminate the observer bias in describing
microdamage as a linear microcrack or diffuse damage. Although the lead-uranyl
acetate based microcomputed tomography technique may not be able to fully resolve
a linear microcrack that is smaller than imaging resolution, pure linear microcracks
are rarely found in bone. A microdamage zone develops around linear microcracks
during growth. The zone allows the lead uranyl acetate to permeate into a wider area
(Fig. 3) [29] and makes it possible to detect microdamage using microcomputed
tomography.
3 The Effect of Aging and Disease on Microdamage

in Cortical and Cancellous Bone
The formation of microdamage is observed in both cancellous and cortical bone.

Previous work on the detection of linear microcracks and diffuse damage in human
vertebral cancellous bone [18, 30, 31] and recent observations of these damage
morphologies in human cortical bone [8] indicate that both cortical and cancellous
tissues form linear microcracks and diffuse damage in vivo.
Fig. 3 A backscattered scanning electron microscopy image illustrating the presence of lead-
uranyl acetate in a linear microcrack and a neighboring region with diffuse damage. Reprinted
with permission from Elsevier [26]
Alterations in microdamage with aging can result from several factors including
deterioration of bone remodeling. Remodeling of bone allows for the removal or
repair of microdamage [3, 30]. There is strong evidence that microdamage initiates
the remodeling process for repair [3, 32]. For instance, application of loading to
produce controlled amounts of microdamage stimulated intracortical remodeling
in rats, which typically do not remodel [30, 33]. Microdamage can also inflict
injury to the osteocytic network or disrupt their nutritional transport, in turn
activating bone remodeling via apoptosis of osteocytes [34]. However, it has been
shown that osteocytes decrease significantly with aging and loss of osteocytes is
associated with accumulation of microdamage [35]. Hence, increasing age may
lead to accumulation of microdamage due to reduction in the ability of bone to
detect and repair damage. Recent studies have also shown that with increasing age
(or due to disease), the efficacy of the bone remodeling processes deteriorates and
thus, microdamage accumulates faster than the remodeling process can repair
these damaged areas [5, 36].
Fig. 4 Microdamage morphology is characterized by DS/DV, which is the surface to volume

ratio of the damaged region. Higher and lower values of DS/DV ratios indicate the presence of
(left) crack-like and (right) diffuse-like microdamage morphologies, respectively. Reprinted with
permission from Elsevier [27]
Reductions in both intracortical bone remodeling in cortical bone [8] and

surface based remodeling on trabecular surfaces of cancellous bone may alter the
accumulation of microdamage with aging [5, 31]. However, due to the inherent
differences in rates of intracortical and surface based remodeling, it is possible that
microdamage accumulation may be different between cortical and cancellous
bone. Particularly, the high surface-to-volume ratio of cancellous bone and low
surface-to-volume ratio of cortical bone paired with similar rates of bone forma-
tion results in an overall higher metabolic activity and turnover rate in cancellous
bone [37]. Consequently, many investigations show an age-related increase of
microdamage in cortical bone [1, 6], but there are discrepancies between inves-
tigations regarding the accumulation of microdamage with age in cancellous bone.
For example, in cancellous bone, some studies found an age-related increase [18,
31, 38] while others report no relationship between microdamage accumulation
and aging [17, 28, 39]. Because anatomical location and mean ages of populations
considered in these studies vary greatly, we suggest that the discrepancies between
these studies could result from differences in turnover rates that vary with ana-
tomical location [40] and age [41]. Consistent with the above concept, a study in
which bone turnover was suppressed using high-dose bisphosphonates showed an
increase in microcrack density in dog vertebral bone [42].
Although microdamage accumulation with aging has been observed for both
males and females (Fig. 5), the limited available data indicate a steeper age-related
increase in microcrack density and age in females than in males [1]. Thus,
microdamage density is significantly greater in older females than in older
males [6]. It is noteworthy that Norman and Wang’s investigation was based on
Fig. 5 Microcrack density in (left) cortical and (right) cancellous bone shows an exponential
increase with age in both males and females. Reprinted with permission from Elsevier and John
Wiley & Sons [1, 31]
bone specimens from older human donors with an average age of 70.5 and
77.9 years for men and women, respectively [6]. It is unclear whether the elevated
microdamage in the old females in this study reflects greater relative accumulation
in the early post-menopausal years (when there is elevated bone turnover) or in the
later years after post-menopausal bone loss has occurred.
In contrast to linear microcracks, the identification and measurement of diffuse
damage is a relatively new area and its relationship to age, gender, and bone
remodeling is still unclear. For example, Vashishth et al. found that diffuse damage
in cancellous bone is compartmentalized primarily near trabecular surfaces, which
is readily accessible for repair by surface based remodeling [15]. There was no
age-related trend in male or female groups but more diffuse damage was present in
men than in women (age range = 23–96 years) [15]. Since the women in this
study were post-menopausal age, typically associated with high bone turnover
rates [43], the existence of more diffuse damage in males than females could be
due to differences in bone turnover. In contrast to Vashishth et al. [15], Arlot et al.
did not find gender based differences, but detected an age-related accumulation of
diffuse damage in trabecular bone (age range = 54–93 years) [31]. This investi-
gation included a larger proportion of older donors, and the results may be
indicative of changes in bone due to senile osteoporosis.
Studies conducted by the author’s group and others have shown the mechanical
effects of diffuse damage on bone fracture (see next section for details). However,
the biological consequences of diffuse damage including damage initiated
remodeling are largely unknown. To date, only a single study by Bentolila et al.
[30] has examined such a relation between diffuse damage and bone resorption.
Unlike linear microcracks, Bentolila et al. found only a statistically non-significant
trend between diffuse damage and bone resorption in rats. Furthermore, unlike
microcrack initiated osteocyte apoptosis [34], no mechanism for diffuse damage
initiated bone resorption has been reported. Because diffuse damage in aging
human cortical bone decreases with age [8], it is likely that diffuse damage triggers
a biological response for its repair and/or reduction. More studies are needed to
examine bone’s in vivo response to diffuse damage.
4 The Contribution of Microdamage to the Biomechanical

Properties of Aging Bone
Bone accumulates fatigue microdamage in vivo due to cyclic loading from everyday
activities. However, as mentioned previously, microdamage repair may be reduced
with the deterioration of the bone remodeling process, leading to an excessive
accumulation of microdamage. This in vivo microdamage accumulation contributes
to deterioration of bone’s mechanical integrity [2, 44] where accumulated micro-
damage as well as its morphology affect both the elastic and post-yield properties of
bone [10, 45, 46].
In particular, microdamage accumulation in the form of linear microcracks is
correlated to loss of material stiffness, or modulus reduction [46–50]. It has been
shown that bone’s elastic modulus decreases after cyclic loading due to accu-
mulation of microdamage, where the modulus loss has a linear relationship with
the amount of diffuse damage and a quadratic relationship with extent of linear
microcracks [46]. It has also been demonstrated that microdamage accumulates
only above a certain threshold of modulus degradation after approximately 15%
stiffness loss [46]. Bone may undergo significant modulus degradation even before
microcracks are evident, and the mechanical properties of bone can be compro-
mised even before substantial microcrack accumulation can be observed on the
microscopic level [46, 51, 52]. These data suggest that the presence of micro-
damage at the sublamellar level may contribute to the deterioration of bone’s
mechanical properties. Consistent with this notion a recent study shows that,
similar to the loss of whole bone material properties, both nano- and micro-
mechanical properties are significantly lower in damaged bone compared to
controls [50].
Bone’s post-yield and fracture properties are more directly assessed through
variables that describe bone’s fracture resistance including strength, toughness,
and crack propagation parameters. Particularly, literature shows that cortical bone
toughness is negatively associated with microdamage accumulation [44, 53].
There is a two- to three-fold increase in microdamage accumulation under sup-
pressed remodeling by bisphosphonates. This accumulation is associated with
approximately 20% decrease in bone toughness without any changes in bone
strength [54]. In contrast to strength, toughness provides a measure of the amount
of energy bone can absorb per volume before failure, independent of the shape or
size of the bone (see Appendix). Currey et al. showed that microdamage accu-
mulation affects bone toughness more significantly than strength [55].
Crack density, size, and propagation parameters measured during or after
fatigue loading also provide useful information about bone’s fracture resistance
[14, 46, 56]. Literature shows a negative correlation between microdamage density
and fracture toughness [53, 57–59]. It has been proposed that bone regions where
cracks easily initiate but do not propagate are more fracture resistant than bone
regions where cracks cannot easily initiate but propagate quickly once formed
[23, 60, 61]. Consequently, fatigue-resistant materials derive their properties
Fig. 6 A schematic representation of microdamage morphologies and their relationship to an

artificially introduced notch on the a compressive and b tensile sides of specimens subjected to
bending fatigue to the primary and tertiary phases of modulus loss. ‘‘x’’ and ‘‘–’’ represent
diffuse damage and linear microcracks, respectively. Reprinted from permission with Elsevier
[10]
mostly from their resistance to crack propagation rather than initiation only
[62–64].
Thus, in contrast to focusing on microdamage that initiates and accumulates
with age, recent studies have focused more on the ability of bone to resist prop-
agation. In particular, several crack propagation studies have been conducted in
which propagation toughness has been measured using a fracture mechanics
approach [7, 45, 62, 65]. Results show that discrete microcrack formation occurs
behind the tip of a propagating fracture crack (frontal process zone) that dissipates
energy and decelerates the advancing fracture [7, 63]. The frontal process zone
develops into a region of microcracks surrounding the propagating fracture
(process zone wake), and both regions absorb energy during loading and lead to
increased crack growth resistance [29, 45, 62, 63]. Energy dissipation through
microcracking may breed into other toughening mechanisms such as uncracked
ligament formation and crack deflection [65]. It is likely that the formation of
uncracked ligaments involves microdamage that arrests a propagating crack and
initiates a new crack [66]. Hence, microdamage forms during crack propagation
(de novo microdamage) and plays a significant role in determining bone’s
toughness [63]. Consequently, increased bone fragility with age may also be due in
part to bone’s decreased ability to form de novo microdamage. Additional studies
are needed to examine this possibility.
Consistent with the above concepts, it has been shown that under fatigue
loading, cortical bone forms and compartmentalizes microdamage in order to
dissipate energy (Fig. 6) [10]. During the primary phase, diffuse damage formed
on the tensile side while few linear microcracks formed on the compressive side.
Furthermore, specimens notched on the compressive side accumulated more
microdamage near the notch and had high toughness. In contrast, specimens
notched on the tensile side had low toughness since the region was already filled
with diffuse damage. Continued loading of specimens into tertiary phase caused
significant accumulation of linear microcracks on the compressive side. More
importantly, linear microcrack accumulation resulted in further toughness loss if

the fracture crack initiated from the compressive rather than tensile side [10].
Results from these studies illustrate that microdamage formation is an energy
dissipation mechanism. Application of loading during regular/strenuous activities
leads to the formation of in vivo microdamage. This microdamage helps to avert
fracture and is removed by bone remodeling. However, deficiency in remodeling
or age-related changes in the bone matrix (see next section for details) cause
increased bone fragility through inefficient repair and alteration in the magnitude
as well as morphology of microdamage formation.
5 The Effect of Changes in Bone Matrix Quality

on Microdamage
Bone derives its resistance to fracture from collagen deformation [67] and by its
ability to form microdamage [62] and uncracked ligament bridges during crack
propagation [65]. Collagen deformation, microcracking (magnitude and mor-
phology), and uncracked ligaments are the primary toughening mechanisms in
bone and any changes in these mechanisms will influence bone toughness. Several
studies are currently ongoing to establish precise mechanisms by which particular
modifications in bone matrix components (e.g. mineral, collagen, non-collagenous
proteins) affect microdamage formation and cause increased bone fragility with
aging, disease, and/or pharmaceutical treatment. A representative example of
modification in bone collagen and its effect on microdamage formation and bone
fragility is discussed below.
Collagen, a key structural component of bone’s extracellular matrix, undergoes
biochemical changes such as non-enzymatic glycation with aging [68, 69] or bis-
phosphonate-based pharmaceutical treatments for osteoporosis [70, 71]. Non-enzymatic
glycation creates crosslinks within and between collagen fibers that are collec-
tively known as advanced glycation end products (AGEs) [72, 73]. AGEs accu-
mulate with age, and their accumulation deteriorates the mechanical properties
of bone [74–77]. Particularly, non-enzymatic glycation of collagen modifies
bone’s post-yield properties [78] and thus may play a crucial role in skeletal
fragility [72, 79, 80].
In context of toughening mechanisms including collagen deformation and
microcracking, Vashishth et al. [78] and Tang et al. [26] showed that accumulation
of AGEs causes stiffening of the collagen matrix in both cortical and cancellous
bone. Also, AGE accumulation leads to decreased deformation and increased
fracture propensity with aging [26] and bisphosphonate-treatment [71]. Further-
more, in a recent study conducted by Tang and Vashishth, results indicated that
AGEs affect both the morphology and magnitude of microdamage formation. They
found that in vitro ribosylated cancellous bone specimens had increased amounts
of linear microcracks whereas controls had relatively more diffuse damage [27].
This data suggests that less glycated bone is able to dissipate more energy through
diffuse damage formation while more glycated bone is less efficient in energy
dissipation due to increased linear microcrack formation. Thus, modifications of
bone matrix components can alter microdamage based toughening mechanisms
and lead to increased bone fragility.
6 Summary
This chapter provided a review of literature on microdamage and its effect on

bone’s fracture behavior. Applied loading and its interaction with the hierarchy in
bone’s extracellular matrix produces microdamage at various length scales where
microdamage can take two distinct forms (linear microcracks or diffuse damage).
Linear microcracks can coalesce and contribute to a large scale fracture. On the
other hand, the submicroscopic cracks of diffuse damage are self-limiting, and
consequently, this form of microdamage is beneficial to bone’s energy absorption
capability, (i.e. toughness). Microdamage is conventionally detected via
two-dimensional histology methods using basic fuchsin for in vivo microdamage
detection or chelating agents for multi-labeling of in vitro induced microdamage.
However, a new technique incorporating heavy metal staining in conjunction
with microcomputed tomography has been developed for the three-dimensional
detection of microdamage. By means of imaging detection and fracture mechanics
based methods for quantification of bone’s fracture properties, it has been deter-
mined that microdamage formation is an energy dissipation mechanism in bone.
The application of loading during regular/strenuous activities leads to formation of
microdamage in vivo. This microdamage helps to prevent fracture and is removed
by bone remodeling. However, deficiency in remodeling or age-related changes in
bone matrix cause increased bone fragility through inefficient repair and changes
in the magnitude and morphology of microdamage formation.
Acknowledgments NIH grants AR49635, AG20618, and T32GM067545.
Appendix: Measurement of Bone’s Fracture Resistance
As mentioned previously, the fracture properties of bone can be altered drastically

due to microdamage accumulation. These fracture properties include measures for
strength (resistance to permanent deformation) and toughness (resistance to frac-
ture) [7, 81]. In order to measure these properties, early studies on bone fracture
utilized the strength-of-materials approach. The traditional method to measure
strength in bone involves mechanical testing on un-notched specimens. This
technique results in the initiation and propagation of fracture from random
distribution of natural flaws. Evaluation of strength with this method is based
on calculation of work-to-fracture (area under load-deformation curve) or the

modulus of toughness (area under stress–strain curve). However, fracture resis-
tance depends on the presence of preexisting defects (i.e. microdamage) in
addition to the stress or stain applied to bone.
Hence, parallel research introduced a fracture mechanics based approach in
which a controlled crack was induced in the specimen before mechanical testing in
order to study bone’s fracture characteristics. The induced sharp pre-crack func-
tions as the dominant flaw from which the crack initiates. Using a pre-cracked
specimen, toughness at initiation and propagation can be measured. Initiation
toughness (critical stress intensity factor [Kc] or strain energy release rate [Gc])
illustrates the inherent toughness of the bone whereas propagation toughness
(slope of crack growth resistance curve) illustrates bone’s resistance to the prop-
agation of a crack [82, 83]. For instance, compact tension specimens with an
induced chevron notch have been successfully used for investigation of crack
propagation and measurement of bone’s fracture resistance [62, 84]. The fracture
mechanics approach has been recently been modified for application on small
animal bones in order to measure whole bone toughness [82, 83]. Here, pre-
cracked rodent long bones (e.g. femur) are mechanically tested via three-point
bending. To calculate fracture toughness using this method, three-dimensional
images obtained via microcomputed tomography can be utilized to pinpoint the
exact location of the notch made in the bone. A cross-sectional image of the notch
can then be imported into imaging software (e.g. ImageJ) to measure the inner and
outer radii, cortical thickness, and notch angles (initial notch and notch at the
instability region). If we assume that the test specimen can be approximated as an
edge-cracked cylindrical pipe, these measurements as well as the load obtained
during fracture can be incorporated to calculate the fracture toughness using
Eq. (1):
Pc S Ro pffiffiffiffiffiffiffiffiffiffiffiffi
k ¼ Fb pH ð1Þ
pðR4o R4i Þ
where k = fracture toughness, Pc = maximum load (maximum load method) or

load at fracture instability (instability method), S = span length, Ro = outer
radius of cortical shell, Ri = inner radius of cortical shell, Rm = mean radius of
cortical shell, H = half-crack angle at crack initiation (maximum load method) or
half crack angle at fracture instability (instability method), t = cortical thickness,
and Fb = geometrical factor for an edge-cracked cylindrical pipe. The geometrical
factor is computed by Eqs. (2) through (8):
" 2 3 4 #
t H H H H
Fb ¼ 1 þ A b þ Bb þ Cb þDb þEb ð2Þ
2Rm p p p p
Ab ¼ 0:65133 0:5774n 0:3427n2 0:0681n3 ð3Þ
Bb ¼ 1:879 þ 4:795n þ 2:343n2 0:6197n3 ð4Þ

Fig. 7 Microcomputed tomography image of a mouse femur showing cortical thickness and
endosteal and periosteal radii measured at three cross-sectional regions. Reprinted with
permission from Elsevier [83]
Cb ¼ 9:779 38:14n 6:611n2 þ 3:972n3 ð5Þ
Db ¼ 34:56 þ 129:9n þ 50:55n2 þ 3:374n3 ð6Þ
Eb ¼ 30:82 147:69n 78:38n2 15:54n3 ð7Þ

t
n ¼ log ð8Þ
Rm
Comparison between the traditional and contemporary methods [82] shows that
the notched technique produces smaller variations in fracture toughness than the
un-notched method. Additionally, it was found that crack propagation properties
measured from the controlled propagation of a crack due to a pre-crack more
comprehensively captures the fracture behavior of bone [62]. As toughness mea-
surements via this fracture mechanics based approach are dependent on applied
loading, geometrical parameters, and a pre-machined notch (Fig. 7), this method
improves the estimation of bone’s fracture resistance compared to that estimated
by traditional methods [82, 83]. (Note that this differs from conventional methods
in which the contributions of structural and material components to bone strength
are not de-coupled.)
References
1. Schaffler, M.B., Choi, K., Milgrom, C.: Aging and matrix microdamage accumulation in
human compact bone. Bone 17, 521–525 (1995)
2. Burr, D.B., Forwood, M.R., Fyhrie, D.P., Martin, R.B., Schaffler, M.B., Turner, C.H.: Bone
microdamage and skeletal fragility in osteoporotic and stress fractures. J. Bone Miner. Res.
12, 6–15 (1997)
3. Burr, D.B., Martin, R.B., Schaffler, M.B., Radin, E.L.: Bone remodeling in response to in
vivo fatigue microdamage. J. Biomech. 18, 189–200 (1985)
4. Noble, B.: Bone microdamage and cell apoptosis. Eur. Cells Mater. J. 6, 46–55 (2003)
5. Waldorff, E.I., Goldstein, S.A., McCreadie, B.R.: Age-dependent microdamage removal
following mechanically induced microdamage in trabecular bone in vivo. Bone 40, 425–432
(2007)
6. Norman, T.L., Wang, Z.: Microdamage of human cortical bone: incidence and morphology in
long bones. Bone 20, 375–379 (1997)
7. Vashishth, D., Tanner, K.E., Bonfield, W.: Contribution, development and morphology of
microcracking in cortical bone during crack propagation. J. Biomech. 33, 1169–1174 (2000)
8. Diab, T., Vashishth, D.: Morphology, localization and accumulation of in vivo microdamage
in human cortical bone. Bone 40, 612–618 (2007)
9. George, W.T., Vashishth, D.: Damage mechanisms and failure modes of cortical bone under
components of physiological loading. J. Orthop. Res. 23, 1047–1053 (2005)
10. Diab, T., Vashishth, D.: Effects of damage morphology on cortical bone fragility. Bone 37,
96–102 (2005)
11. Diab, T., Condon, K.W., Burr, D.B., Vashishth, D.: Age-related change in the damage
morphology of human cortical bone and its role in bone fragility. Bone 38, 427–431 (2006)
12. Burr, D.B., Stafford, T.: Validity of the bulk-staining technique to separate artifactual from in
vivo bone microdamage. Clin. Orthop. Relat. Res. 260, 305–308 (1990)
13. Lee, T.: Microdamage in osteoporosis, bone quality and remodelling. J. Anat. 203, 159
(2003)
14. Boyce, T.M., Fyhrie, D.P., Glotkowski, M.C., Radin, E.L., Schaffer, M.B.: Damage type and
strain mode associations in human compact bone bending fatigue. J. Orthop. Res. 16,
322–329 (1998)
15. Vashishth, D., Koontz, J., Qiu, S.J., Lundin-Cannon, D., Yeni, Y.N., Schaffler, M.B.,
Fyhrie, D.P.: In vivo diffuse damage in human vertebral trabecular bone. Bone 26, 147–152
(2000)
16. Frost, H.M.: Presence of microscopic cracks in vivo in bone. Henry Ford Hosp. Med. Bulletin
8, 25–35 (1960)
17. Wenzel, T.E., Schaffler, M.B., Fyhrie, D.P.: In vivo trabecular microcracks in human
vertebral bone. Bone 19, 89–95 (1996)
18. Mori, S., Harruff, R., Ambrosius, W., Burr, D.B.: Trabecular bone volume and microdamage
accumulation in femoral heads of women with and without femoral neck fractures. Bone 21,
521–526 (1997)
19. Lee, T.C., Mohsin, S., Taylor, D., Parkesh, R., Gunnlaugsson, T., O’Brien, F.J., Giehl, M.,
Gowin, W.: Detecting microdamage in bone. J. Anat. 203, 161–172 (2003)
20. Lee, T.C., Arthur, T.L., Gibson, L.J., Hayes, W.C.: Sequential labelling of microdamage in
bone using chelating agents. J. Orthop. Res. 18, 322–325 (2000)
21. Burr, D.B., Hooser, M.: Alterations to the en bloc basic fuchsin staining protocol for the
demonstration of microdamage produced in vivo. Bone 17, 431–433 (1995)
22. Frost, H.M., Villanueva, A.R., Roth, H., Stanisavljevic, S.: Tetracycline bone labeling.
J. New Drugs 1, 206–216 (1961)
23. O’Brien, F.J., Taylor, D., Lee, T.C.: Microcrack accumulation at different intervals during
fatigue testing of compact bone. J. Biomech. 36, 973–980 (2003)
24. Kazakia, G.J., Lee, J.J., Singh, M., Bigley, R.F., Martin, R.B., Keaveny, T.M.: Automated
high-resolution three-dimensional fluorescence imaging of large biological specimens.
J. Microsc. 225, 109–117 (2007)
25. Bigley, R.F., Singh, M., Hernandez, C.J., Kazakia, G.J., Martin, R.B., Keaveny, T.M.:
Validity of serial milling-based imaging system for microdamage quantification. Bone2 42,
212–215 (2008)
26. Tang, S.Y., Vashishth, D.: A non-invasive in vitro technique for the three-dimensional
quantification of microdamage in trabecular bone. Bone 40, 1259–1264 (2007)
27. Tang, S.Y., Vashishth, D.: Non-enzymatic glycation alters microdamage formation in human
cancellous bone. Bone 46, 148–154 (2010)
28. Karim, L., Vashishth, D.: Role of trabecular microarchitecture in the formation,
accumulation, and morphology of microdamage in human cancellous bone. J. Orthop. Res.
[Epub ahead of print] (2011)
29. Schaffler, M.B., Pitchford, W.C., Choi, K., Riddle, J.M.: Examination of compact bone
microdamage using back-scattered electron microscopy. Bone 15, 483–488 (1994)
30. Bentolila, V., Boyce, T.M., Fyhrie, D.P., Drumb, R., Skerry, T.M., Schaffler, M.B.:
Intracortical remodeling in adult rat long bones after fatigue loading. Bone 23, 275–281
(1998)
31. Arlot, M.E., Burt-Pichat, B., Roux, J.P., Vashishth, D., Bouxsein, M.L., Delmas, P.D.:
Microarchitecture influences microdamage accumulation in human vertebral trabecular bone.
J. Bone Miner. Res. 23, 1613–1618 (2008)
32. Mori, S., Burr, D.B.: Increased intracortical remodeling following fatigue damage. Bone 14,
103–109 (1993)
33. Hsieh, Y.F., Silva, M.J.: In vivo fatigue loading of the rat ulna induces both bone formation
and resorption and leads to time-related changes in bone mechanical properties and density.
J. Orthop. Res. 20, 764–771 (2002)
34. Verborgt, O., Gibson, G.J., Schaffler, M.B.: Loss of osteocyte integrity in association with
microdamage and bone remodeling after fatigue in vivo. J. Bone Miner. Res. 15, 60–67
(2000)
35. Vashishth, D., Verborgt, O., Divine, G., Schaffler, M.B., Fyhrie, D.P.: Decline in osteocyte
lacunar density in human cortical bone is associated with accumulation of microcracks with
age. Bone 26, 375–380 (2000)
36. Nagaraja, S., Lin, A.S., Guldberg, R.E.: Age-related changes in trabecular bone microdamage
initiation. Bone 40, 973–980 (2007)
37. Parfitt, A.M.: Misconceptions (2): turnover is always higher in cancellous than in cortical
bone. Bone 30, 807–809 (2002)
38. Fazzalari, N.L., Kuliwaba, J.S., Forwood, M.R.: Cancellous bone microdamage in the
proximal femur: influence of age and osteoarthritis on damage morphology and regional
distribution. Bone 31, 697–702 (2002)
39. Chapurlat, R.D., Arlot, M., Burt-Pichat, B., Chavassieux, P., Roux, J.P., Portero-Muzy, N.,
Delmas, P.D.: Microcrack frequency and bone remodeling in postmenopausal osteoporotic
women on long-term bisphosphonates: a bone biopsy study. J. Bone Miner. Res. 22,
1502–1509 (2007)
40. Kimmel, D.B., Jee, W.S.: A quantitative histologic study of bone turnover in young adult
beagles. Anat. Rec. 203, 31–45 (1982)
41. Han, Z.H., Palnitkar, S., Rao, D.S., Nelson, D., Parfitt, A.M.: Effects of ethnicity and age or
menopause on the remodeling and turnover of iliac bone: implications for mechanisms of
bone loss. J. Bone Miner. Res. 12, 498–508 (1997)
42. Hirano, T., Turner, C.H., Forwood, M.R., Johnston, C.C., Burr, D.B.: Does suppression of
bone turnover impair mechanical properties by allowing microdamage accumulation? Bone
27, 13–20 (2000)
43. Riggs, B.L., Melton, L.J.: Involutional osteoporosis. N. Engl. J. Med. 314, 1676–1686 (1986)
44. Zioupos, P.: Accumulation of in vivo fatigue microdamage and its relation to biomechanical
properties in ageing human cortical bone. J. Microsc. 201, 270–278 (2001)
45. Zioupos, P., Currey, J.D.: Changes in stiffness, strength, and toughness of human cortical
bone with age. Bone 22, 57–66 (1998)
46. Burr, D.B., Turner, C.H., Naick, P., Forwood, M.R., Ambrosius, W., Hasan, M.S., Pidaparti, R.:
Does microdamage accumulation affect the mechanical properties of bone? J. Biomech. 31,
337–345 (1998)
47. Keaveny, T.M., Wachtel, E.F., Guo, X.E., Hayes, W.C.: Mechanical behavior of damaged
trabecular bone. J. Biomech. 27, 1309–1318 (1994)
48. Yeh, O.C., Keaveny, T.M.: Relative roles of microdamage and microfracture in the
mechanical behavior of trabecular bone. J. Orthop. Res. 19, 1001–1007 (2001)
49. Schaffler, M.B.: Role of bone turnover in microdamage. Osteoporos. Int. 14, S73–S80 (2003)
50. Macione, J., Kavukcuoglu, N.B., Nesbitt, R.S., Mann, A.B., Guzelsu, N., Kotha, S.P.:
Hierarchies of damage induced loss of mechanical properties in calcified bone after in vivo
fatigue loading of rat ulnae. J. Mech. Behavior Biomedical Mater. 4, 841–848 (2011)
51. Schaffler, M.B., Boyce, T.M., Fyhrie, D.P.: Tissue and matrix failure modes in human
compact bone during tensile fatigue. Trans. Orthop. Res. Soc. 21, 57 (1996)
52. Uthgenannt, B.A., Silva, M.J.: Use of the rat forelimb compression model to create discrete
levels of bone damage in vivo. J. Biomech. 40, 317–324 (2007)
53. Norman, T.L., Yeni, Y.N., Brown, C.U., Wang, Z.: Inflence of microdamage on fracture
toughness of the human femur and tibia. Bone 23, 303–306 (1998)
54. Burr, D.: Microdamage and bone strength. Osteoporos. Int. 14, S67–S72 (2003)
55. Currey, J.D., Brear, K., Zioupos, P.: The effects of ageing and changes in mineral content in
degrading the toughness of human femora. J. Biomech. 29, 257 (1996)
56. Pattin, C., Caler, W., Carter, D.: Cyclic mechanical property degradation during fatigue
loading of cortical bone. J. Biomech. 29, 69–79 (1996)
57. Forwood, M.R., Parker, A.W.: Microdamage in response to repetitive torsional loading in the
rat tibia. Calcif. Tissue Int. 45, 47–53 (1989)
58. Schaffler, M.B., Radin, E.L., Burr, D.B.: Mechanical and morphological effects of strain rate
on fatigue of compact bone. Bone 10, 207–214 (1989)
59. Yeni, Y.N., Fyhrie, D.P.: Fatigue damage-fracture mechanics interaction in cortical bone.
Bone 30, 509–514 (2002)
60. Martin, R.B., Burr, D.B., Sharkey, N.A.: Skeletal Tissue Mechanics. Springer, New York
(2004)
61. Chapurlat, R.D.: Bone microdamage. Osteoporos. Int. 20, 1033–1035 (2009)
62. Vashishth, D., Behiri, J.C., Bonfield, W.: Crack growth resistance in cortical bone: concept of
microcrack toughening. J. Biomech. 30, 763–769 (1997)
63. Vashishth, D., Tanner, K.E., Bonfield, W.: Experimental validation of a microcracking-based
toughening mechanism for cortical bone. J. Biomech. 36, 121–124 (2003)
64. Vashishth, D.: Rising crack-growth-resistance behavior in cortical bone: implications for
toughness measurements. J. Biomech. 37, 943–946 (2004)
65. Nalla, R.K., Kinney, J.H., Ritchie, R.O.: Mechanistic fracture criteria for the failure of human
cortical bone. Nat. Mater. 11, 2 (2003)
66. Hazenberg, J.G., Taylor, D., Lee, T.C.: Mechanisms of short crack growth at constant stress
in bone. Biomaterials 27, 2114–2122 (2006)
67. Thompson, J.B., Kindt, J.H., Drake, B., Hansma, H.G., Morse, D.E., Hansma, P.K.: Bone
indentation recovery time correlates with bone reforming time. Nature 414, 773–776 (2001)
68. Saito, M., Marumo, K., Fujii, K., Ishioka, N.: Single-column high-performance liquid
chromatographic-fluorescence detection of immature, mature, and senescent cross-links of
collagen. Anal. Biochem. 253, 26–32 (1997)
69. Odetti, P., Rossi, S., Monacelli, F., Poggi, A., Cirnigliaro, M., Federici, M., Federici, A.:
Advanced glycation end products and bone loss during aging. Ann. NY. Acad. Sci. 1043,
710–717 (2005)
70. Allen, M.R., Gineyts, E., Leeming, D.J., Burr, D.B., Delmas, P.D.: Bisphosphonates alter
trabecular bone collagen cross-linking and isomerization in beagle dog vertebra. Osteoporos.
Int. 19, 329–337 (2008)
71. Tang, S.Y., Allen, M.R., Phipps, R., Burr, D.B., Vashishth, D.: Changes in non-enzymatic
glycation and its association with altered mechanical properties following 1-year treatment
with risedronate or alendronate. Osteoporos. Int. 20, 887–894 (2009)
72. Paul, R.G., Bailey, A.J.: Glycation of collagen: the basis of its central role in the late
complications of ageing and diabetes. Int. J. Biochem. Cell. Biol. 28, 1297–1310 (1996)
73. Bailey, A.J., Paul, R.G., Knott, L.: Mechanisms of maturation and ageing of collagen. Mech.
Ageing Dev. 106, 1–56 (1998)
74. Verzijl, N., DeGroot, J., Oldehinkel, E., Bank, R.A., Thorpe, S.R., Baynes, J.W., Bayliss, M.T.,
Bijlsma, J.W., Lafeber, F.P., Tekoppele, J.M.: Age-related accumulation of Maillard reaction
products in human articular cartilage collagen. Biochem. J. 350(Pt 2), 381–387 (2000)
75. Wang, X., Shen, X., Li, X., Agrawal, C.M.: Age-related changes in the collagen network and
toughness of bone. Bone 31, 1–7 (2002)
76. Vashishth, D.: Age-dependent biomechanical modifications in bone. Crit. Rev. Eukaryot.
Gene Expr. 15, 343–358 (2005)
77. Tang, S.Y., Zeenath, U., Vashishth, D.: Effects of non-enzymatic glycation on cancellous
bone fragility. Bone 40, 1144–1151 (2007)
78. Vashishth, D., Gibson, G.J., Khoury, J.I., Schaffler, M.B., Kimura, J., Fyhrie, D.P.: Influence
of nonenzymatic glycation on biomechanical properties of cortical bone. Bone 28, 195–201
(2001)
79. Hernandez, C.J., van Der Ham, F., Tang, S.Y., Baumbach, B.M., Hwu, P.B., Sakkee, A.N.,
DeGroot, J., Bank, R.A., Keaveny, T.M.: Trabecular microfracture and the influence of
pyridinium and non-enzymatic glycation-mediated collagen cross-links. Bone 37, 825–832
(2005)
80. Viguet-Carrin, S., Garnero, P., Delmas, P.D.: The role of collagen in bone strength.
Osteoporos. Int. 17, 319–336 (2006)
81. Fazzalari, N.L., Forwood, M.R., Smith, K., Manthey, B.A., Herreen, P.: Assessment of
cancellous bone quality in severe osteoarthrosis: bone mineral density, mechanics, and
microdamage. Bone 22, 381–388 (1998)
82. Ritchie, R.O., Koester, K.J., Ionova, S., Yao, W., Lane, N.E., Ager, J.W.: Measurement of the
toughness of bone: a tutorial with special reference to small animal studies. Bone 43,
798–812 (2008)
83. Vashishth, D.: Small animal bone biomechanics. Bone 43, 794–797 (2008)
84. Bonfield, W., Grynpas, M., Young, R.J.: Crack velocity and the fracture of bone. J. Biomech.
11, 473–479 (1978)
Changes in Cortical Bone Mineral
and Microstructure with Aging
and Osteoporosis
Janardhan Yerramshetty and Ozan Akkus
Abstract Microstructural and nanocompositional changes in bone’s fabric with

age and osteoporosis have not been reviewed as much in-depth as the macroscale
changes. This chapter discusses alterations in the size and shape of mineral
crystallites as a function of aging and, also, as result of osteoporosis. The review
also covers the variations in the microstructure, specifically, porosity, fraction of
osteonal bone and collagen orientation. Besides the average values for these
variables, the variation (i.e. heterogeneity) of these variables is also discussed. The
observed results are discussed within the context of changes in the dynamics of
remodeling with age and osteoporosis. Overall, there are discrepancies in the
literature for a significant portion of the variables reported due to a host of issues
such as the lack of statistical power, large scatter inherent in the biological data,
variation of age-range, gender, anatomical region or the methodology employed to
assess a given variable.
J. Yerramshetty
Department of Orthopaedics and Spine Surgery,
Ganga Hospital, Coimbatore 641043, Tamil Nadu, India
e-mail: jyerram@gmail.com
O. Akkus (&)
Department of Mechanical and Aerospace Engineering,
Case Western Reserve University, Cleveland, OH 44106-7222, USA
e-mail: ozan.akkus@case.edu
O. Akkus
Department of Biomedical Engineering, Case Western Reserve University,
Cleveland, OH 44106-7222, USA
O. Akkus
Department of Orthopaedics, Case Western Reserve University,
Cleveland, OH 44106-7222, USA

DOI: 10.1007/8415_2012_114
Published Online: 8 February 2012
106 J. Yerramshetty and O. Akkus
1 Introduction
Aging takes its toll on the tissues through multiple facets such as regular wear and
tear, cumulative exposure to chemical and physical factors, diet, exercise and other
environmental factors [100]. There is also a hereditary aspect to the rate at which
aging occurs. The net effect of aging is gradual impairment in the function of the
tissue. Even though bone is one of the few tissue systems with self-regeneration
capacity, it is also not immune to aging. The fabric-level changes in mineral and
collagen framework results in organ, and thereby, organism level deterioration of
function, which are discerned as alterations in gait, posture and even macro
anatomy such as shortening of spine due to accumulation of creep deformation in
vertebral bodies. This chapter surveys the existing literature on age-related
changes in bone’s mineral and microstructural aspects. It is important to underline
that the present review does not refer aging as pathology, rather, a natural process
during which fractures do not occur. Studies emerging from such sample sets are
categorized as within the aging context. On the other hand, the body of literature
which screened their sample set for osteoporosis, be it by the way of bone mineral
density measurements or by the way of sample harvested from patients with
osteoporotic fractures, is presented separately in Sect. 4. Another important detail
that affects data interpretation is the distinction between growth and maturation
(taking place between birth and adulthood) versus aging (taking place from
adulthood to end of life). Finally, the review is limited to humans and primates, as
the longevity of life and physiology of bone in humans is not sufficiently reflected
by rodent models.
2 Age-Related Changes in Bone’s Mineral Phase
2.1 Bone Matrix Mineral Content
Bone is mainly composed of mineral, collagen and water, where the mineral
composition resembles carbonated apatite, Ca10(PO4)6-x(OH)2-y(CO3)x+y
(Fig. 1). The mineral phase of bone has often been erroneously referred to as
hydroxyapatite. While the presence of Ca and PO43- ions are common in both, the
presence of OH- ion has been contested, particularly in the recent literature
[73, 121]. The structure of mineral lattice, due to a wide array of non-stoichi-
ometric substitutions, makes it even more divergent from hydroxyapatite.
Therefore, carbonated apatite or bone apatite would be more accurate in referring
to bone’s mineral phase.
Generally, changes in bone mineral are discussed with regard to content, size
and shape of the mineral which are measured using variety of methods. Mineral
content at the organ level is often characterized by bone mineral content (BMC),
or bone mineral density (BMD). Specifically, dual-energy X-ray absorptiometry
Changes in Cortical Bone Mineral and Microstructure 107
Fig. 1 Chemical diagram of

23 unit cell of hydroxyapatite
shown in the 001 plane.
Hydroxyapatite has
hexagonal structure of P63/m
and has a unit cell outlined in
the figure with dimensions of:
a = 9.432 Å, b = 9.432 Å,
and c = 6.881 Å, with
calcium (Ca, gray),
phosphate (P, blue), oxygen
(O, red), and hydroxyl (OH,
orange). (Generated by Jared
Diegmueller [56] using
Crystal Maker software)
(DXA) is the most widely applied technique to quantify BMC/BMD changes with
age or osteoporosis, although the BMD measured by DXA represents areal BMD
(aBMD = BMC/projected area). Quantitative computed tomography (QCT)
allows determination of a volumetric BMD (vBMD), a volumetric measure of
bone mineralization. vBMD is a measure inclusive of porosity (i.e. haversian
canals, vascular channels). On the other hand, degree of mineralization (DOM) or
specific mineralization at the solid tissue level, is measured by methods which
resolve bone below microporosity, such as microradiography, ashing, back scat-
tered electron microscopy or spectroscopy (FTIR, Raman). Therefore, the quantity
of bone mineral and how mineralized the bone matrix are, represent two separate
issues which require multiple methods to be addressed specifically.
During osteogenesis, the initial stages, unmineralized osteoid is deposited. The
osteoid is composed of type I collagen, non-collagenous proteins, proteoglycans
and water. Mineralization occurs with some delay within the continuum of the
osteoid (Fig. 2), presumably by displacing water. This theory is indirectly sup-
ported by the study of Mueller et al. study which reported increased degree of
mineralization in elderly with decreased water content [113].
The osteoid turns into fully mineralized bone during two phases. During
primary mineralization, mineral crystals grow and agglomerate to form bigger
15
14 Scan 1 Scan 2 Scan 4
13
Scan 3
Mineralization
12
11
10
9
8
7
6
0 15 30 45 60 75
Distance from Haversian canal µm
Fig. 2 Mineralization of osteoid within a basic multicellular unit (BMU). Raman microspec-
troscopic maps of mineral matrix ratio are overlayed with the optical image. The cement line is
highlighted by a dashed line. Bone formation, temporally, occurs from left to right and from top
to bottom. New bone is formed at the lower section of the mapped regions (haversian canal
surface) where osteoblasts are active. Mineralization increases sigmoidally with with distance
from the Haversian canal
crystals. This process extends over a few days to reach 60–75% of the minerali-
zation of the osteoid [26, 116]. Secondary mineralization is a much slower process
that proceeds for several months, during which the mineral gradually matures
through the process of crystal growth [79]. Recent studies reported that secondary
mineralization may last longer than previously thought. Raman microspectro-
scopic analysis of primary lamellar bone fragments which survived resorption
indicated that crystal growth can last for several decades [5]. It appears that the
supersaturation of interstitial fluid counters the entropic cost of crystal growth
and the overall thermodynamics favor crystal growth in the very long term.
The repercussions of such long term crystal growth on bone matrix, such as
buildup of residual stresses, are currently unknown.
Mineralization at a specific site reflects the age of the mineral crystals involved;
newly formed regions (secondary osteons) are less mineralized whereas old
regions (primary osteons or lamellae) are highly mineralized [124, 126]. The net
mineralization of the matrix, and accordingly the mean tissue age, is mostly
determined by the rate of remodeling. Unfortunately, there is limited insight on
how the rate of local remodeling changes with age, especially via formal dynamic
bone histomorphometry. The most reliable information was reported by Harold
Frost (refer to Sect. 3 for a more detailed discussion) using human ribs which
showed that the remodeling rate declines between birth and the fourth decade, then
increases up to sixth decade and decreases again thereafter [68]. Rehman et al.
characterized the activation frequency from histomorphometry in iliac crest bone
(including the cortical envelope), where they found an insignificant increasing
trend with age in both males and females [125]. Other studies which analyzed
remodeling related parameters (osteoid surfaces, mineralizing surfaces, bone
formation and resorption rates), mostly reported increase in bone resorption,
especially for females, and a decrease in formation indices [42, 47]. Apart from the
ribs and iliac crest, remodeling rate is not well-investigated other locations, par-
ticularly fracture prone regions such as the femoral neck and the vertebral bodies.
There is no consensus on the progression of the mean DOM with age. Some
reported a decline in the amount of primary lamellar bone and the overall minerali-
zation with age [34, 75, 147, 158]. It has been proposed that accumulation of partially
mineralized osteons due to diminished rate of mineralization might result in regions of
less mineralization in elderly [27, 99, 117]. In contrast, others reported an overall
increase in mineralization in elderly, at least until fifth to seventh decades, followed by
a shallow increase [51, 79, 113, 132, 160]. While the design of these studies do not
include quantification of remodeling dynamics, increasing mineralization with age is
attributed to slower remodeling which in turn provides longer time for mineral growth
and increased mineralization of the osteoid. It must be stated that contradictions on
age-related trends in DOM may be a simple matter of statistical power. Probably the
most comprehensive sample size in investigating DOM changes with age was
included in a study by McCalden et al. who observed no change in mineralization with
age in the femurs of males and females [106]. This is further supported by Yeni et al.,
who also found no change in mineralization in individuals aged 60 years and above,
however tibias in the same study had shown lower mineralization with age [158].
Therefore, the third group of studies imply a homeostasis of mean tissue minerali-
zation with age in which the new bone formation balance against the aging old
compartments, maintaining a constant mean tissue age after the third decade [5].
Techniques such as density fractionation, microradiography and electron
backscattering imaging allow for inferring the distribution of mineralization, i.e.,
assessment of high and low mineralization fractions [18, 25, 126, 132, 140, 150].
High density bone was reported to increase with age at the expense of low density
bone for which the researchers provided diverse interpretations that ranged from
reduced remodeling at specific sites [124, 126] to densely calcified fibrocartilage at
ligament attachment sites [140] and lack of lamellar regions at sites possibly
caused by the formation of woven bone [25] and/or fracture callus due to acci-
dental overloads [36, 101].
Mean mineralization does not reflect the heterogeneity of bone’s composition.
For instance, a set of individuals may have the same mean mineralization, yet the
variation of mineralization around the mean value could be larger for some
individuals, or, the variation may be skewed to one side. Several investigators have
appreciated such heterogeneity and reported on the variability of mineralization
(VOM; referred to by some as bone mineralization density distribution, BMDD)
[46, 50, 75, 115, 134, 135, 160]. VOM is calculated as the ratio of standard
deviation to mean (i.e. the coefficient of variation). For instance, a reduction in
remodeling might result in greater mean mineralization, but also reduced standard
deviation of mineralization (thus, reduced VOM) as demonstrated in bisphosph-
onate treated animals [133]. On the other hand, increased remodeling may increase
the variability of mineralization in the short-term by increasing co-existence of
newly formed bone along with old unresorbed bone.
Reports on age-related changes in VOM vary. Reid et al. noted a trend (no
statistics were reported) of reduced VOM with aging (i.e. tighter distribution around
the mean), based on analysis using BSEM technique on the sixth ribs of thirteen
individuals (age range: 2 months–59 years) [126]. Similarly, Roschger et al.
reported a (non-significant) trend of decreasing VOM with aging in transiliac
biopsies collected from males and females (30–85 years) [132]. Moreover, Raman
spectroscopic analysis on cortical bone of femurs obtained near the proximal region
(minor trochanter area) from 16 males (52–85 years) indicated a linear decrease in
the VOM with age [160]. However, in the second study Roschger et al. performed
multi factor analysis of cancellous bone from various sites and individuals and found
no significant change in VOM with age (25–95 years), site, gender or ethnicity,
concluding that VOM was ‘‘essentially constant in healthy adult humans’’ [134].
In contrast, Goldman et al. and Bloebaum et al. observed increased VOM in older
subjects versus younger subjects [18, 75]. Goldman et al. analyzed sections using
BSEM on mid shaft femurs of 40 individuals, which were grouped into three ages
(25–44, 45–65, and 65+ years), while Bloebaum et al. analyzed using a combination
of ashing and BSEM on cancellous region (neck region) of proximal femur grouped
into only two age groups (17–35 and 76–95 years). More recently, an asymmetry in
the spread of mineralization towards higher or lower mineralization is reported by
Yerramshetty et al. using skewness, a statistical measure. The results indicated that
the distribution of mineralization become more increasingly skewed towards higher
mineralization with aging [160].
It is clear that the reported changes in both DOM and VOM with age widely
vary in the literature. The discrepancies remain unexplained, but, in general, they
may be attributed to lack of statistical power in the face of the scatter in biological
data, variation of age-range, gender, anatomical region or the methodology
employed to assess mineralization.
2.2 Size, Shape and Stoichiometry of Mineral Nanocrystals
Mineral crystals orient along the long axis of collagen fibers [66]. Bone mineral
crystals are plate shaped [73, 149] and they are poorly crystalline (compared to
Fig. 3 a Transmission electron micrographic images of mineral nanoplatelets extracted from

bovine bone indicates that mineral crystallites are platelet shaped. Needle-like crystals are not
discernable from TEM images [90]. b As opposed to TEM, atomic force microscopy can provide
information on the three-dimensional topography of mineral crystallites. Note that the orthogonal
axes do not scale proportionally, such that the height axis is enlarged to bring out crystal
thickness which is otherwise about one tenth of crystal length/width. (Courtesy of Dr. Steven
J. Eppell, Case Western Reserve University. See articles [58, 149] for further information on
AFM-based assessment of crystallite morphology.)
Fig. 4 Illustration of
carbonate substitutions
in bone apatite
pure apatite, for example) due to high concentrations of impurities [156]. The
average dimensions of crystals are 9 9 6 9 2 nm (Fig. 3) [149]. The increased
surface area due to nanoplatelet conformation increases non-stoichiometric sub-
stitution. TEM sections of platelets usually reveal ‘needle’ like crystals and early
literature misinterpreted such sections which concluded that mineral crystallites
are rod-shaped. A similar misinterpretation also emerged from earlier X-ray
diffraction studies which did not essentially apply form factors to X-ray diffraction
patterns to obtain the actual shape. The spatial averaging of platelet morphology
thus resulted in a rod-like perception. Most researchers investigate changes along
the longer or c-axis of the crystal, aligned along collagen fibrils, and sometimes
along the a-axis or the thickness, perpendicular to c-axis.
The most common substitution found in the bone lattice is of carbonate
ions. Bone mineral is initially formed as an amorphous calcium phosphate, which
with maturity transforms into a poorly crystalline form of carbonated apatite
[16, 127, 128]. The carbonate ion is distributed or substituted in three different
locations (Fig. 4). When a phosphate ion (PO43-) present in an apatite crystal is
replaced by a carbonate ion (CO32-), it is called as type-B carbonate substitution,

which is the most abundant carbonate species in bone. Carbonate ions occupying
hydroxide ion (OH-) sites are designated as type-A substitution. Carbonate
ions loosely bound on the crystal surface are called labile carbonate [16, 120, 122,
127–129, 156]. Sometimes PO43- ions are substituted by acid phosphate ions
(HPO42-), which are found in a labile non-apatitic environment as well [19, 22]. The
mechanisms by which these substitutions occur are not well known. Type-A carbonate
substitution, when attempted synthetically, required high temperature (*1000°C),
whereas synthetic type-B carbonation occured within a temperature range of
50–100°C [156]. Therefore, carbonate ion substitutions do not seem to be favored
thermodynamically and may be driven biologically via proteins. In support of this
concept, Dziak and Akkus demonstrated that carbonation of in vitro crystals grown in
cell cultures can be modified by charged polypeptides; therefore, non-collagenous
proteins may be involved in carbonate substitution [57].
Carbonated apatites have a significant role in cellular metabolic activities, such
as, when there is an increase in systemic metabolic acidosis due to declined renal
function or other age-related factors. It was hypothesized that systemic acidity
increases bone resorption, releasing carbonate (CO32-) and hydrogen phosphate
(HPO32-) ions to buffer serum pH [6, 37, 38, 65]. Other than the metabolic
activities, carbonate ions, particularly type B substitution, distorts the shape of
mineral crystallites [54, 120] inducing microstrains. The changes were such that
crystallite size, inferred from crystallinity parameter, were reported to decrease
along a-axis and increase along c-axis [81, 93–95].
Only a few studies have investigated age-related changes in type-B carbonation
of human bone. Akkus et al. used Raman spectroscopy on regions of bone that
survived remodeling for decades. Such isolated locations displayed significantly
less type-B carbonation (measured as a ratio of carbonate to phosphate peak)
compared to younger bone tissue. However, after normalization of the carbonate
peak by the amide (collagen’s backbone conformation) peak, they found no
change in type-B carbonate, which led them to conclude that the reduction in type-
B carbonate to phosphate ratio was due to an increase in phosphate content over
time rather than a decline in carbonate substitution [5]. Later, a study by
Yerramshetty et al. on human femoral cortical bone revealed a significant increase
with age in mean carbonate to phosphate ratio among an older population
(52–85 years) [160]. Additionally, a significant reduction was reported in the
standard deviation of carbonation with age in the medial quadrant of femoral
cortex, and the overall distribution skewed towards higher levels of carbonation.
Higher levels and reduced variation of carbonation imply more immature
crystallites in aged individuals, as carbonation generally reduces with mineral
maturity and perfection. Relative amounts of labile carbonate, measured using
density fractionation technique, obtained from the bone of a single individual
increased with density suggesting that labile carbonates are principally formed
during later stages of mineralization. Similar results were also observed in syn-
thetic apatites [128]. Handschin et al. employed chemical analysis to monitor age
related changes in the total carbonate content [30, 81] in human iliac crest samples
Fig. 5 Data from a 71 year 0.27

old male. The line shows an
inversely related linear
relationship between 0.24
Carbonation
carbonation and crystallinity,
R2 = 50.2%, P \ 0.05 [160] 0.21
0.18
0.15
0.05 0.055 0.06 0.065
Crystallinity
and found increase in carbonate content with age similar to others, who also found
that mineral crystallinity, obtained from X-ray diffraction, decreased with age.
Raman spectroscopic analysis on a sample set consisting of human femoral cor-
tical bone from individuals of different ages led to a conclusion that there is an
inverse relationship between carbonation and crystallinity (Fig. 5) [160].
Crystallinity parameter is a composite indicator of crystal size and/or lattice
perfection and it reflects the overall maturity of the crystal [4, 5, 10, 67, 81, 105,
108, 128, 129]. On the other hand, this parameter cannot resolve between the
constituent effectors. Smaller crystals with fewer imperfections may technically
have similar crystallinity as larger crystals with more imperfections. Since it is
believed that onset of crystal nucleation occurs in an amourphous state, it is
expected that crystal size and lattice perfection are positively associated. X-ray
diffraction and spectroscopic techniques have been widely used to characterize
crystallinity. In a systematic approach on a large sample set, Handschin et al.,
assessed association between age and crystallinity, which is generally affected by
molecular order (perfection), domain size and lattice strains [81]. Table 1 lists
some of the studies on humans that had reported age related trends. Simmons
et al.’s conclusion based on no change in crystallite size with degree of miner-
alization, which normally should increase contradicts the general theory of growth
of mineral crystals as reported by other researchers.
Akkus et al.’s observation supported their hypothesis that crystals mature
during the secondary mineralization process to a saturation level and then maintain
a constant crystallinity for the rest of the crystal’s life [5]. They further added that
when bone is homogenized, mature mineral crystals get mixed with smaller
crystallites, which are mostly found in new secondary osteons formed due to bone
turnover, resulting in a constant crystallinity in homogenized bone as observed
in the study [5]. Similar to Akkus et al.’s study (after 40 years of age), no change
was observed with age in crystallinity in femoral cortex of elderly individuals
(52–85 years) despite an increase in carbonation [160]. Stagnant crystallinity in
the face of increased carbonation was explained by a scenario where increase in
crystal sizes (increase crystallinity) offset the effect of increased carbonation
(decrease crystallinity).
114
Table 1 Studies related to age associated changes in crystallinity of bone

Author and year Technique Changes in crystallinity with age Conclusions
Handschin and Stern (1995) [81] X-ray diffraction Along c-axis: increased until third Incorporation of carbonate ions
decade and remained constant and reduction in chloride ions
after that decrease perfection and
Along a-axis: No changes until increase crystal size
third decade but slightly
increased after that.
(0–90 years)
Holden et al. (1995) [83] X-ray diffraction Increased with age (1–97 years) Poorly crystalline state when new
bone is laid down and is
converted to more crystalline
form with age
Simmons et al. (1991) [140] X-ray diffraction No change between age groups From formation to maturation
(20–85 years) mineral crystals are of standard
size
Akkus et al. (2003) [5] Raman spectroscopy Increased until 40 years and Crystal maturation occurs during a
remained protracted secondary
constant there after mineralization period and
(17–73 years) remain stagnant
Yerramshetty et al. (2006) [160] Raman spectroscopy No change with age (52–82 years) Increase in crystal sizes are
compensated by increased
carbonate substitution
J. Yerramshetty and O. Akkus
3 Aging and Microstructure of Cortical Bone
Modeling of bone results in changes in the size and shape of bone, while
remodeling turns over the internal bone volume. The rate of modeling is largely
reduced after skeletal maturation whereas remodeling is a life-long process [101].
Remodeling occurs in three stages (activation, resorption and formation) that
replaces old with new bone and serves three main functions; first, to balance
essential minerals within the serum, second to build skeletal strength by adapting
to mechanical needs, and finally to repair microdamage formed during daily
activities that prevents fracture risk [33, 101]. The first function can be accom-
plished by removing bone indiscriminately from any location. On the other hand,
the latter two functions require site-specific remodeling [33]. Therefore, remod-
eling is categorized into targeted (site-specific) and non-targeted forms [33, 103,
112, 116]. A haversian canal is formed at the center of newly formed secondary
osteons, and a reversal line, also called as cement line, is created at the transition
region between old and newly formed bone, which is often used to differentiate
secondary osteons from primary osteons [101].
Remodeling has been shown to increase following menopause [53, 123].
Remodeling rate is believed to decrease during aging [68]. The most reliable means
to infer remodeling is histomorphometry; however, biopsy requirement and inac-
cessibility of sample collection from points of interest (femoral neck, vertebral
bodies) limit this approach. Nonetheless, earlier mathematical methods [68] predict
remodeling rate using resorption spaces and refilling basic multicellular units
(BMUs) identified by osteoid seams. Net remodeling representing the average
remodeling that occurred during the life-time of the individual is termed as previous
remodeling in this chapter and it was quantified by examining the amount of section
occupied by secondary osteons and their remnants [82, 104, 114, 145]. Remodeling
rate is also assessed based on biochemical bone turnover markers (alkaline phos-
phatase, osteocalcin, tartrate-resistant acid phosphate, collagen telopeptide markers)
from serum or urine [2, 17]. The limitation of this method is that it is a systemic
measure. Remodeling estimated but such tests reflect current remodeling, i.e., at a
particular instant in the individual’s life.
Frost used biopsies from human ribs [68] to estimate remodeling dynamics via
the activation frequency variable (a.k.a. bone turnover rate–number of osteons/
mm2/year). Ribs were chosen because of the ease in obtaining biopsy, and they are
subjected to high remodeling rates under continuous cyclic loading due to
breathing. Frost administered tetracycline stains twice within a period of 10 days,
which labeled mineralizing bone twice [68]. The author observed that activation
frequency declined to a lifetime minimum level at 35 years of age which slightly
increased in late 50s, and declined again towards 90s [68]. These data indicated
that bone was rapidly converting into secondary osteonal form and maintained at
relatively undermineralized levels at younger ages, followed by increased
mineralization during adulthood, when systemic factors like hormonal changes
have reduced remodeling to its minimum level. The increase in remodeling in the
late 50s could happen possibly for mechanical and/or other metabolic reasons
[101]. Surprisingly, except for a couple of histomorphometric studies on iliac crest
[125], where upward trend with age was noticed, it is unknown if the same age-
related patterns hold true for other bones.
Frost [69] developed an empirical algorithm, which was capable of estimating
missing osteons in histological sections to calculate activation frequency––infor-
mation that can normally be obtained only from in vivo labeling [1, 32, 35, 63, 82,
92, 145]. Stout et al. (human ribs), Havill et al. (macaca mulatto’s femur), Frank
et al. (dog humerus) and Lees et al. (cynomolgus monkey’s iliac bone), using the
algorithm developed by Frost, were able to demonstrate that activation frequency
reduced with age. However, studies using biochemical turnover markers, observed
no change in remodeling activity with age [2]. Remodeling is ultimately associated
with porosity, fraction of osteonal bone and to some extent collagen orientation
which are discussed hence forth.
3.1 Porosity
Haversian canals, resorption spaces, vascular channels, marrow spaces, osteocyte

lacunae and canaliculi contribute to porosity (voids) in bone [102]. The small
dimensions of osteocyte level porosity (lacunae and canaliculi) limited its evalu-
ation; however recent advances in staining and imaging techniques made it pos-
sible to analyze these tiny interstitial fluid spaces [45]. Haversian canals are the
spaces formed during remodeling periods, which contain vessels with nutrients and
other precursor cells necessary for BMUs. Osteoblasts form bone, leading from the
outer edges until they encounter vessels. Porosity increases when bone formation
lags relative to bone resorption [102]. Porosity in cortical bone is in the range of
5–10%, much lower than in cancellous bone (75–95%) [49]. Increased cortical
porosity is caused by increased number of haversian canals, greater number of
resorption spaces, or by incompletely filled osteons (and thus increased diameter
of haversian canals). Moreover, focal accumulation of resorption spaces, such as
those observed close to the endosteal surfaces, may be indicative of a diseased
state such as osteoporosis [61, 88, 102]. In addition, cortical porosity is associated
with local stress levels from habitual loading such that anterior and posterior
quadrants of femoral mid shaft, which experience lower stresses, have greater
porosity than medial and lateral quadrants [103, 154].
Increased volume fraction of cortical porosity with aging has been reported by
many studies [61, 103, 106, 140, 143, 150, 161]. The rate of change in porosity
with age is reported to vary anatomically. Femoral neck and intertrochanteric
cortical porosity increase at a greater rate when compared to the diaphysis, which
was speculated to be caused by targeted remodeling [150]. In another study,
increased porosity with aging was found in the mid-neck region while lesser
increases were observed in the subcapital level and no age-related changes in the
trochanteric region. Moreover, no difference in porosity was observed between
subcapital, mid-neck and trochanteric regions of femoral neck in the younger age
group, but differences were seen in the older age groups with greatest void volume
being present in the mid-neck region [25]. Bell et al. demonstrated the interesting
concept of super-osteons, which are spatially clustered remodeling osteons [15].
This merging of osteonal systems to form large cortical cavities can have dele-
terious effects on bone strength, but super-osteons were found to be independent of
age and gender [15]. Age-related increases in porosity and the differential response
over anatomical locations points out that change in mechanical milieu, particularly
age-related reduction in physical activity, may play a key role in site-specific
increase in porosity. It is also important to appreciate that such site specific
changes are likely taking place over a background of general increase in porosity
due to endocrine factors. The relative roles of endocrine and mechanical factors in
modulating age-related cortical porosity are not well understood.
3.2 Osteonal Microanatomy
During remodeling, osteoclasts dig tunnels of around 200 lm diameter and osteoblasts
fill into form cylindrical tubular structures known as secondary osteons or Haversian
systems. Primary bone starts to convert into secondary at a very young age and
becomes mostly secondary by adult age [68]. Changes in bone microanatomy during
remaining life are mixed.
Kerley reported increasing densities of osteonal fragments and secondary osteons
with age in femurs, tibias and fibulas from both genders (birth to 94 years of age)
[89]. However in human femurs, Wang et al. observed no change in the number of
secondary osteons per unit area with age, which were clustered into three age groups
between 19 and 89 years [155]. It was observed that the percentage of bony area in
each osteon decreased [106, 155], indicating bigger central haversian canals in older
individuals. Overall size of osteons (including haversian canal) also increased in
elderly, but their number remained constant during individual’s lifespan [106].
Conversely Britz et al. found reduced overall osteon size with age, measured as
osteonal area and osteonal diameter, in femurs of both males and females [28], with
female femurs consisting significantly smaller osteons than males. The authors noted
a negative relationship between weight and osteon size. Based on Martin’s study
[103], Britz et al. speculated that reduced osteoclastic activity might be a reason for
reduced osteon size in older bones [28]. In a study on mid-diaphysis of male femurs,
Zioupos measured the fraction of secondary osteonal area adjusted for haversian
canal area and found increased fractional area with age even after correction for the
vascular canal size [161].
Boyce et al. found a greater number of osteons per mm2 in the trochanteric
region than in other regions of the femoral neck. However, the osteon density did
not change with age in both trochanteric and subcapital regions, but the mid-neck
region had greater number of osteons/mm2 in the older age group compared to the
younger age group [25]. Osteon population density (intact ? fragmented)
increased with age in the primates, similar to humans. However, no change was
observed in osteonal area or percentage of osteonal bone, likely due to reduced
activation frequency and bone formation rate [82].
3.3 Collagen Orientation
At the microscopic level, two kinds of bone are evident, lamellar and woven bone.
Lamellar bone, by definition, is slowly formed and highly organized bone con-
sisting of parallel layers of collagen fibrils and mineral crystals with long axis
oriented along the collagen fibers [9]. In contrast, woven bone is quickly formed
and poorly organized bone with randomly arranged fibers and mineral crystals
[101]. Von Ebner [152] was the first to suggest about the orientation of collagen
fibers that appeared to change between successive lamellae of an osteon [74].
During the 60s and 70s, Ascenzi and Bonucci [8] and Frasca et al. [64] used
birefringence to study collagen orientation, where bone specimens are illuminated
and viewed through polarizing filters. Because of longitudinal and transverse
collagen fiber orientation, dark and bright fields are produced by the rotation of
plane of polarized light [8, 64, 153]. Later, Giraud-Guille, using transmission
electron microscopy (TEM), suggested two kinds of collagen organization. The
first kind of organization is analogous to orthogonal plywood, where within each
lamella fibers orient parallel, but the next lamella change direction by 90° at the
interface. Depending on the bone section considered, dark and bright images
are seen corresponding to transverse and longitudinal orientations, respectively, in
the TEM. The second kind is similar to twisted plywood, closely relating to the
‘alternating’ or ‘intermediate’ orientation suggested by Ascenzi, where collagen
fibers continuously change direction as a series of nested arcs [72]. Giraud-Guille
concluded that human compact bone consisted of both kinds. Contradicting the
classical models, Marotti et al. suggested another kind of organization where
osteons consist of collagen rich (densely packed) and collagen poor (loosely
packed) areas, producing bright and dark appearances under polarized light,
respectively [98]. However, Ascenzi’s and Bonucci’s model is widely accepted
and used to describe the birefringence pattern in cortical bone.
Spiraling collagen fibers were found in younger individuals when compared to
the old by Amprino et al., who suggested that elderly individuals tended to have
more transversely oriented collagen fibers [151]. Conversely, Smith demonstrated
light osteons (transverse) converted to dark osteons (longitudinal) with aging in
human femora and tibia [141]. It is still unclear what contributes to the orientation
of collagen fibers. Rearrangement of crystallites and fibers within a calcified and
matured osteons is highly unlikely but mechanical loading may be the determining
factor for fiber orientation at the time of osteon formation [151]. In support, more
transverse orientation was observed in newly formed osteons in older individuals
whereas newly formed osteons in younger individuals tended to have longitudinal
orientation.
Goldman et al. performed a study of the femoral mid-shaft where the whole-bone
cross section was covered comprehensively. The amount of transverse collagen
fibers among males decreased between the younger and middle age groups, and
increased between the middle and older groups, whereas females showed this trend
only along the endosteal ring of the cortical cross-section. Similar to Vincentelli
et al., Goldman group opined that the mechanical environment in the mid-shaft of the
femur was a possible reason for the changes in collagen orientation [76]. Long ago,
Kuntscher [91] associated tensile and compressive stress and strain patterns to dif-
ferent quadrants of long bones. Using this theory, Portigliatti et al. came to the
conclusion that strain patterns have a significant role in fiber orientation in the middle
third of femoral diaphysis. Fibers were aligned more longitudinally in the lateral and
anterior quadrants (predominantly loaded in tension), compared to transversely in
the medial and posterior quadrants (predominantly loaded in compression) [124].
Others have reported similar observations [131, 146].
4 Changes in Bone’s Mineral Phase and Microstructure

During Osteoporosis
In cortical regions, osteoporosis is mainly characterized by enlargement of haver-

sian canals and cortical thinning caused by the net loss of bone from the endosteal
surface. In cancellous regions, thin and perforated trabeculae are found during
osteoporosis, caused by the removal of bone from plates and struts. DXA-based
BMD has been the single most widely accepted tool for diagnosis of osteoporosis,
despite many questions about the specificity and sensitivity of using BMD
alone. Bone strength measured ex vivo correlates moderately well to BMD
[43, 52, 78, 137], and fracture risk increases with lower BMD [139, 144]. Yet in
most studies BMD fails to fully discriminate between people suffering fractures and
those who are not [50, 115, 136]. Therefore, diagnosis solely based on T and/or Z
scores of BMD readings lacks fidelity [115]. Association of BMD with fracture is
site specific, for instance, fractured group had significantly lower spine BMD but not
hip BMD [77]. Similarly in the same study, hip BMD of cortical bone, but not
cancellous bone, had significant association with fracture risk [77]. However,
Holzer et al. found marginal difference between the cancellous and cortical con-
tribution to bone strength in femoral neck region [84]. Advances in structural level
assessment of microstructures (using microCT, high resolution CT, etc.) along with
macro (BMD using radiographs, DXA, etc.) may improve fracture risk assessment
[71]. In conjunction with the above concerns and shortcomings of BMD in estab-
lishing a reliable estimator for bone strength, the focus was increased on aspects like
architecture, whole bone morphology and bone matrix; all of which are unified
under the term ‘bone quality’. This section will expand on mineral crystals and
micro-structural aspects of bone quality in the osteoporotic population.
BMD is a non-invasive macroscopic measure that normalizes bone mineral
content (BMC) measured over a specific area at the organ level. BMD inherently
includes effects of porosity as well, which increases with bone loss and is well
related to skeletal failures [13, 14, 55]. BMD is distinct from degree of mineral-
ization (DOM), which is quantitated microscopically and measured at tissue level
(refer to Sect. 2.1) and is often represented as a ratio against collagen matrix.
Earlier studies observed higher degree of mineralization in trabecular bone of
osteoporotic individuals (low BMD) suggesting that despite bone loss, individuals
can still have higher DOM [7]. Bone mineralization was also found to correlate
highly with strength [62] implying bone quality measures may complement BMD
in predicting fracture risk.
Degree of mineralization was observed to be greater in individuals with hip
fractures [142, 157]. Wu et al. suggested that low turnover may possibly be
responsible for the increase in DOM, yet, a lower bone turnover assumption does
not align well with the reduction in bone mass in osteoporotic individuals or the
positive association between bone turnover and fracture risk. An alternative
explanation of increasing average mineralization in the face of increasing
resorption and less bone is based on a relative increase in more mineralized
interstitial bone with greater turnover of selective osteons [55]. In other words,
bone of younger tissue age is preferentially resorbed and the bone that remains
tends to have older tissue age and thus be more mineralized. In contrast to above
mentioned studies, reduced mineralization is reported in osteoporotic humans [96]
and overectomized monkeys [70], a good model for post-menopausal osteoporotic
women [87]. Loveridge reported less mineralization in the femoral neck regions of
patients with hip fractures when compared to controls and they could not establish
any relationship between mineralization and remodeling dynamics. In their
opinion vitamin D deficiency might be the potential cause for decreased miner-
alization in fractured individuals because involvement of vitamin D deficiency was
earlier implied in hip fractures [60], which is believed to increase bone turnover
and decline mineralization rate, resulting in fast removal and slow maturation of
bone mineral. Other reasons like genetic defects in collagen and osteocyte apop-
tosis were also considered in interpreting their results. However, the samples had
no known genetic defects and were neither measured for vitamin D nor analyzed
for osteocytes.
Boskey et al. has pointed out that osteoporosis can occur by high turnover
(greater than normal resorption rate) or low turnover (lower than normal formation
rate) [23]. High turnover regime tends to shift the balance to lower mineralization
and lower crystallinity, whereas the low turnover regime may result in a relatively
greater mineralization and mineral maturity. Therefore, while both regimes result
in lower bone mass there may be differing outcomes of osteoporosis in terms of
matrix mineralization. It is clear that remodeling dynamics shapes the degree of
mineralization, besides the rate of crystal growth, and such associations remain to
be investigated. Biochemical markers can be used to distinguish high and low
turnover osteoporosis as a systemic indicator of remodeling [44, 53, 107]. Yet,
occurrence of fractures in specific anatomical locations indicates the need for site
specific means to determine resorption and formation dynamics. As of yet, such
analyses cannot be conducted non-invasively and mostly depend on assessment of
invasive biopsies by spectroscopic techniques (FTIR, Raman) or imaging tech-

niques (microCT, high resolution CT).
From the above discussions, data with respect to mineralization and osteopo-
rosis are contradictory (Table 2). Such disparity suggests that both lower and
higher mineralization may associate with osteoporotic bone loss. A case study
conducted by Ciarelli et al. reported bimodal mineralization distribution in frac-
tured cases (i.e. greater standard deviation than normal cases) such that both the
lower and upper peak regions were associated with fractured cases [46]. High and
low mineralization levels each have their own drawbacks in terms of biome-
chanical properties at tissue level. Highly mineralized specimens tend to be brittle
and absorb less amount of energy before fracture, whereas lower mineralized
specimens have lower stiffness. The disparity on the trends of mineralization in
osteoporosis may also be a result of excessive spatial heterogeneity, study design
(age range included, criterion on what constitutes osteoporosis, fractured versus
osteoporotic samples) and simple lack of power due to limited availability of
biopsies of fractured samples. Importantly, most studies do not report BMD and
DOM simultaneously; therefore, the relation of matrix level mineral content to
BMD is unknown.
Irrespective of various observations in average mineralization levels (Table 2),
most investigators agreed that the statistical spread of mineral content within the
field of interest either narrowed, indicating a more homogeneous distribution
(similar aged minerals) [23], or found no change especially in trabecular bone [24].
With advanced technology and more research on distribution parameters, heter-
ogeneity of mineralization could provide an interesting complementary parameter
to distinguish osteoporotic from normal individuals.
Mineral crystal status (size, crystallinity and substitutions) has also been
investigated in osteoporotic individuals. There is a lack of consensus in the
literature regarding the state of mineral crystals. A group of studies reported that
the osteoporotic bones have longer and/or more perfect mineral crystals compared
to normal individuals, using techniques like spectroscopy and X-ray diffraction
[22, 23, 29, 48, 70, 118, 148]. Others reported no change in crystal size [77, 97],
while some found decreased crystal size in individuals with osteoporosis [11, 80].
As individual mineral crystals grow in size, the mineral content is expected to
increase, therefore, changes in mineral crystal’s size to some extent reflect degree
of mineralization [21]. However, in some cases average crystallinity observed over
an area in osteoporotic and fractured individuals was found to be greater than
normal, despite average mineralization being lower [22, 23]. It is speculated that if
resorption were to act more efficiently and rapidly on smaller crystals, then the
larger ones would be left behind. Because of statistical averaging, mineralization
on the whole may seem less, but large crystals left behind may increase average
crystallinity. Similarly a low formation and/or higher maturation rates can lead to
higher crystallinity. For above reasons, a change in statistical spread was also
noticed in such a way that mineral crystal’s size distribution in both cortical and
cancellous bone narrowed and shifted to higher levels in osteoporotic patients
[21, 23]. Similar to BMD and mineralization, crystallinity was also suggested [41]
Table 2 Mineralization levels found in osteoporotic and fractured cases (relative to case con-
trols) and various opinions on possible causes and phases affected
Mineralization level Possible causes Affected turnover phases
Higher mineralization Lower turnover [142, 157] Lower resorption, higher maturation
Higher turnover [55] Higher resorption, lower formation
Lower mineralization Lower turnover [23] Lower resorption, lower formation
Higher turnover [23, 96] Higher resorption, lower maturation,
same or increased formation
and later found to correlate with mechanical properties at tissue level [159] and
was significantly associated with bone fracture in a multiple logistic regression
model evaluated for cancellous bone [77]. Crystallinity was also implicated in
crack initiation and subsequent growth due to decreased deformability before
failure [31, 67, 159]. Both mechanistic models [3, 86] and experimental studies
[159] found that long crystals contribute to increased stiffness and strength, but at
the expense of ductility.
Repercussions of crystal size in osteoporotic individuals are not clear yet, but as
can be seen, specimens with large crystals potentially lack ductility and in turn they
may contribute to failure during falls. On the other hand, bone tissue with smaller
crystals lack stiffness and may fail progressively due to compromised weight bearing
capacity over prolonged periods, such as the creep observed in vertebrae. As an
example, Camacho et al. observed significantly large crystals in cancellous bone
compared to cortical bone in the vertebra of normal individuals [39].
Substitutions by ions like carbonate, fluoride, chloride and other heavy metal
ions in mineral crystals could also affect bone’s quality by influencing crystal size
and/or perfection. Decreased levels of carbonate content were reported in osteo-
porotic female Eskimos compared to healthy males [148]. Carbonate molecule
plays a significant role in the dissolution of mineral and resorption of bone
[10, 95]. In general, it is considered that new bone contains small crystals that will
have large quantities of carbonate content thus making bone imperfect and easily
removable or dissolvable during the process of resorption; as bone mineral matures
carbonate content decreases. Thompson et al. stated that in the case of osteopo-
rosis, in which bone turnover increases, younger bone was resorbed leaving behind
the mature bone that contained low carbonate content [148]. Huang et al. com-
pared site (cortical versus cancellous) and type (A, B and labile) comparisons
between normal and ovariectomized monkeys [85]. Type-A and type-B levels
increased in the cortical bone of ovariectomized sample, while type-A increased
and type-B decreased in trabecular bone and labile carbonate was found in lower
levels in both cortical and cancellous bone. Similar perception was also shared by
Boskey et al. and Gadeleta et al., that carbonate accumulation decreases with
mineral maturation, however Boskey group did not observe any differences
between normal and osteoporotic patients [23] while Gadeleta et al. observed
increased levels of type-B carbonation in osteoporotic individuals and overiec-
tomized cynomolgus monkeys [70].
Various studies assessed treatment strategies of osteoporosis to counter bone loss,

particularly the bisphosphonates and anabolic agents. In most cases, bishphospho-
nates succeeded in increasing bone density [59, 111] and degree of mineralization
[20, 135] by prolonging secondary mineralization. Hormone replacement therapy
(HRT) was used for the treatment and prevention of osteoporosis, which improved
mineral content and the statistical distribution of the population of crystal sizes in
postmenopausal women [118]. A decrease in the average crystal size was observed
in early postmenopausal women treated with HRT [118]. Fluoride therapy was one
of the treatments that directly affected mineral properties, especially crystal size,
along with mineral density [12]. Though crystal length was increased, crystal width
was decreased, which would reduce the ability of bone tissue to bear high loads [40].
That may be one reason why fluoride treatment failed to reduce the incidence of hip
fractures [130]. In addition, the mineral crystal’s interaction with organic matter
may have been disturbed by the altered crystal size in fluoride therapy [109].
Intermittent parathyroid hormone treatment (PTH) increased formation of new
bone matrix and mineral crystals [119]. This conclusion was based on the
observation of shifts in mineral/matrix ratio and crystallinity towards lower values
in both animals [138] and humans [110]. Other bone quality properties like
compositional changes in bone minerals caused by anabolic agents were also
investigated. Overiectomized rats [19] and monkeys [70] treated with bisphos-
phonates and monitored using infrared spectroscopy demonstrated no change in
carbonate and acid phosphate content. However cancellous bone in dogs treated
with calcitonin, which inhibits osteoclasts, exhibited increased levels of carbonate
which possibly may have happened due to delayed mineralization.
5 Conclusion
Remodeling holds the key in terms of determining the mean tissue age.
A mechanically competent tissue structure depends on temporally and spatially
orchestrated remodeling process. A delicate balance between activation, resorption
and formation processes is essential to turn over older tissue fragments and repair
in vivo damage accumulation. The relative rates of these three processes can
potentially alter in many different combinations during aging and osteoporosis.
It may be why there is a lack of consensus in the literature on compositional,
architectural and microstructural changes during aging and osteoporosis. The
number of cells involved in the basic multicellular unit and their metabolic
capacity may change with age and osteoporosis as well. Another variable at play is
the rate of mineralization of the osteoid. Finally, aging of unresorbed moieties via
prolonged crystal growth and non-enzymatic crosslinks further complicates the
picture. A complete reliance on biopsies in terms of investigating these factors
reliably curbs the progress of research. Novel technologies reporting on matrix
quality and remodeling dynamics in situ and non-invasively (or minimally inva-
sively) would be a quantum-leap in the field.
A specific challenge in understanding matrix level alterations in osteoporosis is

based on the uncertainty of defining what comprises an osteoporotic sample. If the
criterion is a BMD measurement, BMD by and itself is not a reliable indicator of
fracture risk. Another approach would be biopsy collection from individuals with
past fracture history and in such circumstances the compounding effects of past
fracture cannot be adjusted easily.
Acknowledgments This material is based upon work supported by the National Science
Foundation under Grant No. 0620061.
References
1. Abbott, S., Trinkaus, E., Burr, D.B.: Dynamic bone remodeling in later pleistocene fossil
hominids. Am. J. Phys. Anthropol. 99(4), 585–601 (1996)
2. Agnusdei, D., Civitelli, R., Camporeale, A., Parisi, G., Gennari, L., Nardi, P., Gennari, C.:
Age-related decline of bone mass and intestinal calcium absorption in normal males. Calcif.
Tissue Int. 63(3), 197–201 (1998)
3. Akkus, O.: Elastic deformation of mineralized collagen fibrils: an equivalent inclusion
based composite model. J. Biomech. Eng. 127(3), 383–390 (2005)
4. Akkus, O., Adar, F., Schaffler, M.B.: Age-related changes in physicochemical properties of
mineral crystals are related to impaired mechanical function of cortical bone. Bone 34(3),
443–453 (2004)
5. Akkus, O., Polyakova-Akkus, A., Adar, F., Schaffler, M.B.: Aging of microstructural
compartments in human compact bone. J. Bone Miner. Res. 18(6), 1012–1019 (2003)
6. Arnett, T.: Regulation of bone cell function by acid-base balance. Proc. Nutr. Soc. 62(2),
511–520 (2003)
7. Arnold, J.S., Bartley, M.H., Tont, S.A., Jenkins, D.P.: Skeletal changes in aging and disease.
Clin. Orthop. Relat. Res. 49, 17–38 (1966)
8. Ascenzi, A., Bonucci, E.: The tensile properties of single osteons. Anat. Rec. 158(4),
375–386 (1967)
9. Ascenzi, M.G., Ascenzi, A., Benvenuti, A., Burghammer, M., Panzavolta, S., Bigi, A.:
Structural differences between ‘‘dark’’ and ‘‘bright’’ isolated human osteonic lamellae.
J. Struct. Biol. 141(1), 22–33 (2003)
10. Baig, A.A., Fox, J.L., Young, R.A., Wang, Z., Hsu, J., Higuchi, W.I., Chhettry, A., Zhuang, H.,
Otsuka, M.: Relationships among carbonated apatite solubility, crystallite size, and
microstrain parameters. Calcif. Tissue Int. 64(5), 437–449 (1999)
11. Baud, C.A., Pouezat, J.A., Tochon-Danguy, H.J.: Quantitative analysis of amorphous and
crystalline bone tissue mineral in women with osteoporosis. Calcif. Tissue Res. 21(Suppl),
452–456 (1976)
12. Baud, C.A., Very, J.M., Courvoisier, B.: Biophysical study of bone mineral in biopsies of
osteoporotic patients before and after long-term treatment with fluoride. Bone 9(6), 361–365
(1988)
13. Bell, K.L., Loveridge, N., Power, J., Garrahan, N., Meggitt, B.F., Reeve, J.: Regional
differences in cortical porosity in the fractured femoral neck. Bone 24(1), 57–64 (1999)
14. Bell, K.L., Loveridge, N., Power, J., Garrahan, N., Stanton, M., Lunt, M., Meggitt, B.F.,
Reeve, J.: Structure of the femoral neck in hip fracture: cortical bone loss in the
inferoanterior to superoposterior axis. J. Bone Miner. Res. 14(1), 111–119 (1999)
15. Bell, K.L., Loveridge, N., Reeve, J., Thomas, C.D., Feik, S.A., Clement, J.G.: Super-osteons
(remodeling clusters) in the cortex of the femoral shaft: influence of age and gender. Anat.
Rec. 264(4), 378–386 (2001)
16. Biltz, R.M., Pellegrino, E.D.: The nature of bone carbonate. Clin. Orthop. Relat. Res. 129,
279–292 (1977)
17. Blain, H., Vuillemin, A., Blain, A., Guillemin, F., De Talance, N., Doucet, B., Jeandel, C.:
Age-related femoral bone loss in men: evidence for hyperparathyroidism and insulin-like
growth factor-1 deficiency. J. Gerontol. A. Biol. Sci. Med. Sci. 59(12), 1285–1289 (2004)
18. Bloebaum, R.D., Lundeen, G.A., Shea, J.E., Whitaker, E.L.: Age-related mineralization
heterogeneity changes in trabecular bone of the proximal femur. Anat. Rec. A. Discov. Mol.
Cell. Evol. Biol. 281(2), 1296–1302 (2004)
19. Bohic, S., Rey, C., Legrand, A., Sfihi, H., Rohanizadeh, R., Martel, C., Barbier, A., Daculsi, G.:
Characterization of the trabecular rat bone mineral: effect of ovariectomy and bisphosphonate
treatment. Bone 26(4), 341–348 (2000)
20. Boivin, G.Y., Chavassieux, P.M., Santora, A.C., Yates, J., Meunier, P.J.: Alendronate
increases bone strength by increasing the mean degree of mineralization of bone tissue in
osteoporotic women. Bone 27(5), 687–694 (2000)
21. Boskey, A.: Bone mineral crystal size. Osteoporos. Int. 14(Supplement 5), 16–21 (2003)
22. Boskey, A., Mendelsohn, R.: Infrared analysis of bone in health and disease. J. Biomed.
Opt. 10(3), 031102 (2005)
23. Boskey, A.L., DiCarlo, E., Paschalis, E., West, P., Mendelsohn, R.: Comparison of mineral
quality and quantity in iliac crest biopsies from high- and low-turnover osteoporosis: an
FT-IR microspectroscopic investigation. Osteoporos. Int. 16(12), 2031–2038 (2005)
24. Boskey, A.L., Spevak, L., Weinstein, R.S.: Spectroscopic markers of bone quality in
alendronate-treated postmenopausal women. Osteoporos. Int. 20(5), 793–800 (2009)
25. Boyce, T.M., Bloebaum, R.D.: Cortical aging differences and fracture implications for the
human femoral neck. Bone 14(5), 769–778 (1993)
26. Boyde, A., Compston, J.E., Reeve, J., Bell, K.L., Noble, B.S., Jones, S.J., Loveridge, N.:
Effect of estrogen suppression on the mineralization density of iliac crest biopsies in young
women as assessed by backscattered electron imaging. Bone 22(3), 241–250 (1998)
27. Boyde, A., Jones, S.J.: Aspects of anatomy and development of bone: the nm, lm and mm
hierarchy. Adv. Organ Biol. 5A, 3–44 (1998)
28. Britz, H.M., Thomas, C.D.L., Clement, J.G., Cooper, D.M.L.: The relation of femoral
osteon geometry to age, sex, height and weight. Bone 45(1), 77–83 (2009). doi:10.1016/
j.bone.2009.03.654
29. Burnell, J.M., Baylink, D.J., Chestnut, C.H., 3rd, Mathews, M.W., Teubner, E.J.: Bone
matrix and mineral abnormalities in postmenopausal osteoporosis. Metab. Clin. Exp.
31(11), 1113–1120 (1982)
30. Burnell, J.M., Teubner, E.J., Miller, A.G.: Normal maturational changes in bone matrix,
mineral, and crystal size in the rat. Calcif. Tissue Int. 31(1), 13–19 (1980)
31. Burr, D.: Microdamage and bone strength. Osteoporos. Int. 14(Suppl 5), S67–S72 (2003)
32. Burr, D.B.: Estimated intracortical bone turnover in the femur of growing macaques:
implications for their use as models in skeletal pathology. Anat. Rec. 232(2), 180–189 (1992)
33. Burr, D.B.: Targeted and nontargeted remodeling. Bone 30(1), 2–4 (2002)
34. Burr, D.B., Martin, R.B.: The effects of composition, structure and age on the torsional
properties of the human radius. J. Biomech. 16(8), 603–608 (1983)
35. Burr, D.B., Ruff, C.B., Thompson, D.D.: Patterns of skeletal histologic change through
time: comparison of an archaic native American population with modern populations. Anat.
Rec. 226(3), 307–313 (1990)
36. Burr, D.B., Schaffler, M.B., Yang, K.H., Lukoschek, M., Sivaneri, N., Blaha, J.D.,
Radin, E.L.: Skeletal change in response to altered strain environments: is woven bone a
response to elevated strain? Bone 10(3), 223–233 (1989)
37. Bushinsky, D.A.: Acid-base imbalance and the skeleton. Eur. J. Nutr. 40(5), 238–244 (2001)
38. Bushinsky, D.A., Lam, B.C., Nespeca, R., Sessler, N.E., Grynpas, M.D.: Decreased bone
carbonate content in response to metabolic, but not respiratory, acidosis. Am. J. Physiol.
265(4 Pt 2), F530–F536 (1993)
39. Camacho, N.P., Rinnerthaler, S., Paschalis, E.P., Mendelsohn, R., Boskey, A.L., Fratzl, P.:
Complementary information on bone ultrastructure from scanning small angle X-ray
scattering and fourier-transform infrared microspectroscopy. Bone 25(3), 287–293 (1999)
40. Chachra, D., Turner, C.H., Dunipace, A.J., Grynpas, M.D.: The effect of fluoride treatment
on bone mineral in rabbits. Calcif. Tissue Int. 64(4), 345–351 (1999)
41. Chatterji, S., Wall, J.C., Jeffery, J.W.: Age-related changes in the orientation and particle
size of the mineral phase in human femoral cortical bone. Calcif. Tissue Int. 33(6), 567–574
(1981)
42. Chavassieux, P., Meunier, P.J.: Histomorphometric approach of bone loss in men. Calcif.
Tissue Int. 69(4), 209–213 (2001)
43. Cheng, X.G., Lowet, G., Boonen, S., Nicholson, P.H., Brys, P., Nijs, J., Dequeker, J.:
Assessment of the strength of proximal femur in vitro: relationship to femoral bone mineral
density and femoral geometry. Bone 20(3), 213–218 (1997)
44. Christiansen, P.: The skeleton in primary hyperparathyroidism: a review focusing on bone
remodeling, structure, mass, and fracture. Apmis 109, 5–52 (2001)
45. Ciani, C., Doty, S.B., Fritton, S.R.: An effective histological staining process to visualize
bone interstitial fluid space using confocal microscopy. Bone 44(5), 1015–1017 (2009).
doi:10.1016/j.bone.2009.01.376
46. Ciarelli, T.E., Fyhrie, D.P., Parfitt, A.M.: Effects of vertebral bone fragility and bone
formation rate on the mineralization levels of cancellous bone from white females. Bone
32(3), 311–315 (2003). doi:10.1016/s8756-3282(02)00975-4
47. Clarke, B.L., Ebeling, P.R., Jones, J.D., Wahner, H.W., O’Fallon, W.M., Riggs, B.L.,
Fitzpatrick, L.A.: Changes in quantitative bone histomorphometry in aging healthy men.
J. Clin. Endocrinol. Metab. 81(6), 2264–2270 (1996)
48. Cohen, L., Kitzes, R.: Infrared spectroscopy and magnesium content of bone mineral in
osteoporotic women. Israel J. Med. Sci. 17(12), 1123–1125 (1981)
49. Cowin, S.C.: Bone Mechanics Handbook, 2nd edn. CRC Press, Boca Raton (2001)
50. Cummings, S.R.: Are patients with hip fractures more osteoporotic? Review of the
evidence. Am. J. Med. 78(3), 487–494 (1985)
51. Currey, J.D., Brear, K., Zioupos, P.: The effects of ageing and changes in mineral content in
degrading the toughness of human femora. J. Biomech. 29(2), 257–260 (1996)
52. D’Amelio, P., Rossi, P., Isaia, G., Lollino, N., Castoldi, F., Girardo, M., Dettoni, F., Sattin, F.,
Delise, M., Bignardi, C.: Bone mineral density and singh index predict bone mechanical
properties of human femur. Connect. Tissue Res. 49(2), 99–104 (2008). doi:10.1080/
03008200801913940
53. De Leo, V., Ditto, A., la Marca, A., Lanzetta, D., Massafra, C., Morgante, G.: Bone mineral
density and biochemical markers of bone turnover in peri- and postmenopausal women.
Calcif. Tissue Int. 66(4), 263–267 (2000)
54. Demul, F.F.M., Otto, C., Greve, J., Arends, J., Tenbosch, J.J.: Calculation of the Raman line
broadening on carbonation in synthetic hydroxyapatite. J. Raman Spectrosc. 19(1), 13–21
(1988)
55. Dickenson, R.P., Hutton, W.C., Stott, J.R.: The mechanical properties of bone in
osteoporosis. J. Bone Joint Surg. Br. 63-B(2), 233–238 (1981)
56. Diegmueller, J.J.: The Effects of Polyelectrolytic Peptides on the Size and Shape of
Hydroxyapatite Nanocrystals. Purdue University, West Lafayette (2008)
57. Dziak, K.L., Akkus, O.: Effects of polyelectrolytic peptides on the quality of mineral
crystals grown in vitro. J. Bone Miner. Metab. 26(6), 569–575 (2008). doi:10.1007/s00774-
008-0869-x
58. Eppell, S.J., Tong, W., Katz, J.L., Kuhn, L., Glimcher, M.J.: Shape and size of isolated
bone mineralites measured using atomic force microscopy. J. Orthop. Res. 19(6),
1027–1034 (2001)
59. Eriksen, E.F., Melsen, F., Sod, E., Barton, I., Chines, A.: Effects of long-term risedronate on
bone quality and bone turnover in women with postmenopausal osteoporosis. Bone 31(5),
620–625 (2002)
60. Faccini, J.M., Extonsmith, A.N., Boyde, A.: Disorders of bone and fracture of femoral-
neck––evaluation of computer image analysis in diagnosis. Lancet 1(7969), 1089–1092
(1976)
61. Feik, S.A., Thomas, C.D., Bruns, R., Clement, J.G.: Regional variations in cortical
modeling in the femoral mid-shaft: sex and age differences. Am. J. Phys. Anthropol. 112(2),
191–205 (2000)
62. Follet, H., Boivin, G., Rumelhart, C., Meunier, P.J.: The degree of mineralization is a
determinant of bone strength: a study on human calcanei. Bone 34(5), 783–789 (2004).
doi:10.1016/j.bone.2003.12.012
63. Frank, J.D., Ryan, M., Kalscheur, V.L., Ruaux-Mason, C.P., Hozak, R.R., Muir, P.: Aging
and accumulation of microdamage in canine bone. Bone 30(1), 201–206 (2002)
64. Frasca, P., Harper, R.A., Katz, J.L.: Collagen fiber orientations in human secondary osteons.
Acta. Anat. (Basel) 98(1), 1–13 (1977)
65. Frassetto, L., Morris Jr, R.C., Sellmeyer, D.E., Todd, K., Sebastian, A.: Diet, evolution and
aging–the pathophysiologic effects of the post-agricultural inversion of the potassium-to-
sodium and base-to-chloride ratios in the human diet. Eur. J. Nutr. 40(5), 200–213 (2001)
66. Fratzl, P., Fratzl-Zelman, N., Klaushofer, K., Vogl, G., Koller, K.: Nucleation and growth of
mineral crystals in bone studied by small-angle X-ray scattering. Calcif. Tissue Int. 48(6),
407–413 (1991)
67. Freeman, J.J., Wopenka, B., Silva, M.J., Pasteris, J.D.: Raman spectroscopic detection of
changes in bioapatite in mouse femora as a function of age and in vitro fluoride treatment.
68. Frost, H.M.: Tetracycline-based histological analysis of bone remodeling. Calcif. Tissue
Res. 3(3), 211–237 (1969)
69. Frost, H.M.: Secondary osteon population-densities––an algorithm for estimating the
missing osteons. Yearb. Phys. Anthropol. 30, 239–254 (1987)
70. Gadeleta, S.J., Boskey, A.L., Paschalis, E., Carlson, C., Menschik, F., Baldini, T., Peterson,
M., Rimnac, C.M.: A physical, chemical, and mechanical study of lumbar vertebrae from
normal, ovariectomized, and nandrolone decanoate-treated cynomolgus monkeys (Macaca
fascicularis). Bone 27(4), 541–550 (2000)
71. Genant, H.K., Engelke, K., Prevrhal, S.: Advanced CT bone imaging in osteoporosis.
Rheumatology 47, 9–16 (2008). doi:10.1093/rheumatology/ken180
72. Giraud-Guille, M.M.: Twisted plywood architecture of collagen fibrils in human compact
bone osteons. Calcif. Tissue Int. 42(3), 167–180 (1988)
73. Glimcher, M.J.: Bone: nature of the calcium phosphate crystals and cellular, structural, and
physical chemical mechanisms in their formation. Rev. Mineral. Geochem. 64(1), 223–282
(2006). doi:10.2138/rmg.2006.64.8
74. Goldman, H.M.: Histocomposition and geomety at the human mid-shaft femur.
Dissertation, University of New York (2001)
75. Goldman, H.M., Bromage, T.G., Boyde, A., Thomas, C.D., Clement, J.G.: Intrapopulation
variability in mineralization density at the human femoral mid-shaft. J. Anat. 203(2),
243–255 (2003)
76. Goldman, H.M., Bromage, T.G., Thomas, C.D., Clement, J.G.: Preferred collagen fiber
orientation in the human mid-shaft femur. Anat. Rec. A. Discov. Mol. Cell Evol. Biol.
272(1), 434–445 (2003)
77. Gourion-Arsiquaud, S., Faibish, D., Myers, E., Spevak, L., Compston, J., Hodsman, A.,
Shane, E., Recker, R.R., Boskey, E.R., Boskey, A.L.: Use of FTIR spectroscopic imaging to
identify parameters associated with fragility fracture. J. Bone Miner. Res. 24(9), 1565–1571
(2009)
78. Gregory, J.S., Stewart, A., Undrill, P.E., Reid, D.M., Aspden, R.M.: Bone shape, structure,
and density as determinants of osteoporotic hip fracture––a pilot study investigating the
combination of risk factors. Invest. Radiol. 40(9), 591–597 (2005)
79. Grynpas, M.: Age and disease-related changes in the mineral of bone. Calcif. Tissue Int.
53(Suppl 1), S57–S64 (1993)
80. Grynpas, M.D., Holmyard, D.: Changes in quality of bone mineral on aging and in disease.
Scanning Microsc. 2(2), 1045–1054 (1988)
81. Handschin, R.G., Stern, W.B.: X-ray diffraction studies on the lattice perfection of human
bone apatite (Crista iliaca). Bone 16(4 Suppl), 355S–363S (1995)
82. Havill, L.M.: Osteon remodeling dynamics in Macaca mulatta: normal variation with regard
to age, sex, and skeletal maturity. Calcif. Tissue Int. 74(1), 95–102 (2004)
83. Holden, J.L., Clement, J.G., Phakey, P.P.: Age and temperature related changes to the
ultrastructure and composition of human bone mineral. J. Bone Miner. Res. 10(9),
1400–1409 (1995)
84. Holzer, G., von Skrbensky, G., Holzer, L.A., Pichl, W.: Hip fractures and the contribution of
cortical versus trabecular bone to femoral neck strength. J. Bone Miner. Res. 24(3), 468–474
(2009). doi:10.1359/jbmr.081108
85. Huang, R.Y., Miller, L.M., Carlson, C.S., Chance, M.R.: In situ chemistry of osteoporosis
revealed by synchrotron infrared microspectroscopy. Bone 33(4), 514–521 (2003)
86. Jager, I., Fratzl, P.: Mineralized collagen fibrils: a mechanical model with a staggered
arrangement of mineral particles. Biophys. J. 79(4), 1737–1746 (2000)
87. Jerome, C.P., Carlson, C.S., Jayo, M.J., Register, T.C., Weaver, D.S., Lees, C.J.,
Adams, M.R.: Histomorphometric and mineral density fractionation studies of lumbar
vertebrae of intact and ovariectomized (OVX) monkeys. Bone Miner. 26(3), 275–278 (1994)
88. Jordan, G.R., Loveridge, N., Bell, K.L., Power, J., Rushton, N., Reeve, J.: Spatial clustering
of remodeling osteons in the femoral neck cortex: a cause of weakness in hip fracture? Bone
26(3), 305–313 (2000)
89. Kerley, E.R.: The microscopic determination of age in human bone. Am. J. Phys.
Anthropol. 23(2), 149–163 (1965)
90. Kim, H.M., Rey, C., Glimcher, M.J.: Isolation of calcium-phosphate crystals of bone by
non-aqueous methods at low temperature. J. Bone Miner. Res. 10(10), 1589–1601 (1995)
91. Kuntscher, G.: Die Darstellung des Kraftflusses im Knochen. Zentbl. Chir. 61, 2130–2136 (1934)
92. Lees, C.J., Ramsay, H.: Histomorphometry and bone biomarkers in cynomolgus females: a
study in young, mature, and old monkeys. Bone 24(1), 25–28 (1999)
93. LeGeros, R.Z., Kijkowska, R., Bautista, C., LeGeros, J.P.: Synergistic effects of magnesium
and carbonate on properties of biological and synthetic apatites. Connect. Tissue Res.
33(1–3), 203–209 (1995)
94. LeGeros, R.Z., Trautz, O.R., Klein, E., LeGeros, J.P.: Two types of carbonate substitution in
the apatite structure. Experientia 25(1), 5–7 (1969)
95. LeGeros, R.Z., Tung, M.S.: Chemical stability of carbonate- and fluoride-containing
apatites. Caries. Res. 17(5), 419–429 (1983)
96. Loveridge, N., Power, J., Reeve, J., Boyde, A.: Bone mineralization density and femoral
neck fragility. Bone 35(4), 929–941 (2004)
97. Lundon, K., Dumitriu, M., Grynpas, M.: The long-term effect of ovariectomy on the quality
and quantity of cancellous bone in young macaques. Bone Miner. 24(2), 135–149 (1994)
98. Marotti, G.: A new theory of bone lamellation. Calcif. Tissue Int. 53(Suppl 1), S47–S55
(1993), discussion S56
99. Marotti, G., Favia, A., Zallone, A.Z.: Quantitative analysis on the rate of secondary bone
mineralization. Calcif. Tissue Res. 10(1), 67–81 (1972)
100. Martin, G.: Biology of aging. In: Goldman, L., Ausiello, D. (eds.) Cecil Medicine, 23rd edn.
Saunders Elsevier, Philadelphia (2007)
101. Martin, R.B., Burr, D.B., Sharkey, N.A.: Skeletal Tissue Mechanics. Springer, New York (1998)
102. Martin, R.B., Ishida, J.: The relative effects of collagen fiber orientation, porosity, density,
and mineralization on bone strength. J. Biomech. 22(5), 419–426 (1989)
103. Martin, R.B., Pickett, J.C., Zinaich, S.: Studies of skeletal remodeling in aging men. Clin.
Orthop. Relat. Res. 149, 268–282 (1980)
104. Martin, R.B., Stover, S.M., Gibson, V.A., Gibeling, J.C., Griffin, L.V.: In vitro fatigue
behavior of the equine third metacarpus: remodeling and microcrack damage analysis.
J. Orthop. Res. 14(5), 794–801 (1996)
105. Matsushima, N., Hikichi, K.: Age changes in the crystallinity of bone mineral and in the
disorder of its crystal. Biochim. Biophys. Acta. 992(2), 155–159 (1989)
106. McCalden, R.W., McGeough, J.A., Barker, M.B., Court-Brown, C.M.: Age-related changes
in the tensile properties of cortical bone. The relative importance of changes in porosity,
mineralization, and microstructure. J. Bone Joint Surg. Am. 75(8), 1193–1205 (1993)
107. Mehl, B., Delling, G., Schlindwein, I., Heilmann, P., Voia, C., Ziegler, R., Nawroth, P.,
Kasperk, C.: Do markers of bone metabolism reflect the presence of a high- or low-turnover
state of bone metabolism? Med. Klin. 97(10), 588–594 (2002). doi:10.1007/s00063-002-
1199-8
108. Mendelsohn, R., Paschalis, E.P., Sherman, P.J., Boskey, A.L.: IR microscopic imaging of
pathological states and fracture healing of bone. Appl. Spectrosc. 54(8), 1183–1191 (2000)
109. Misof, B., Roschger, P., Klaushofer, K.: Evaluations for Yerramshetty JS & Akkus O Bone
42(3), 476–482 (2008). Faculty of 1000 Medicine. http://www.f1000medicine.com/article/
id/1453957
110. Misof, B.M., Roschger, P., Cosman, F., Kurland, E.S., Tesch, W., Messmer, P.,
Dempster, D.W., Nieves, J., Shane, E., Fratzl, P., Klaushofer, K., Bilezikian, J., Lindsay, R.:
Effects of intermittent parathyroid hormone administration on bone mineralization density in
iliac crest biopsies from patients with osteoporosis: a paired study before and after treatment.
J. Clin. Endocrinol. Metab. 88(3), 1150–1156 (2003)
111. Monier-Faugere, M.C., Geng, Z., Paschalis, E.P., Qi, Q., Arnala, I., Bauss, F., Boskey, A.L.,
Malluche, H.H.: Intermittent and continuous administration of the bisphosphonate
ibandronate in ovariohysterectomized beagle dogs: effects on bone morphometry and
mineral properties. J. Bone Miner. Res. 14(10), 1768–1778 (1999)
112. Mori, S., Burr, D.B.: Increased intracortical remodeling following fatigue damage. Bone
14(2), 103–109 (1993)
113. Mueller, K.H., Trias, A., Ray, R.D.: Bone density and composition. Age-related and
pathological changes in water and mineral content. J. Bone Joint Surg. Am. 48(1), 140–148
(1966)
114. Mulhern, D.M., Van Gerven, D.P.: Patterns of femoral bone remodeling dynamics in a
Medieval Nubian population. Am. J. Phys. Anthropol. 104(1), 133–146 (1997)
115. Nielsen, S.P.: The fallacy of BMD: a critical review of the diagnostic use of dual X-ray
absorptiometry. Clin. Rheumatol. 19(3), 174–183 (2000)
116. Parfitt, A.M.: Bone age, mineral density, and fatigue damage. Calcif. Tissue Int. 53(Suppl 1),
S82–S85 (1993), discussion S85–S86
117. Parfitt, A.M., Han, Z.H., Palnitkar, S., Rao, D.S., Shih, M.S., Nelson, D.: Effects of ethnicity
and age or menopause on osteoblast function, bone mineralization, and osteoid
accumulation in iliac bone. J. Bone Miner. Res. 12(11), 1864–1873 (1997)
118. Paschalis, E.P., Boskey, A.L., Kassem, M., Eriksen, E.F.: Effect of hormone replacement
therapy on bone quality in early postmenopausal women. J. Bone Miner. Res. 18(6),
955–959 (2003)
119. Paschalis, E.P., Burr, D.B., Mendelsohn, R., Hock, J.M., Boskey, A.L.: Bone mineral and
collagen quality in humeri of ovariectomized cynomolgus monkeys given rhPTH(1–34) for
18 months. J. Bone Miner. Res. 18(4), 769–775 (2003)
120. Paschalis, E.P., DiCarlo, E., Betts, F., Sherman, P., Mendelsohn, R., Boskey, A.L.: FTIR
microspectroscopic analysis of human osteonal bone. Calcif. Tissue Int. 59(6), 480–487
(1996)
121. Pasteris, J.D., Wopenka, B., Freeman, J.J., Rogers, K., Valsami-Jones, E., van der Houwen,
J.A.M., Silva, M.J.: Lack of OH in nanocrystalline apatite as a function of degree of atomic
order: implications for bone and biomaterials. Biomaterials 25(2), 229–238 (2004)
122. Penel, G., Leroy, G., Rey, C., Bres, E.: MicroRaman spectral study of the PO4 and CO3
vibrational modes in synthetic and biological apatites. Calcif. Tissue Int. 63(6), 475–481
(1998)
123. Podenphant, J., Johansen, J.S., Thomsen, K., Riis, B.J., Leth, A., Christiansen, C.: Bone
turnover in spinal osteoporosis. J. Bone Miner. Res. 2(6), 497–503 (1987)
124. Portigliatti Barbos, M., Bianco, P., Ascenzi, A.: Distribution of osteonic and interstitial
components in the human femoral shaft with reference to structure, calcification and
mechanical properties. Acta. Anat. (Basel) 115(2), 178–186 (1983)
125. Rehman, M.T., Hoyland, J.A., Denton, J., Freemont, A.J.: Age related histomorphometric
changes in bone in normal British men and women. J. Clin. Pathol. 47(6), 529–534 (1994)
126. Reid, S.A., Boyde, A.: Changes in the mineral density distribution in human bone with age:
image analysis using backscattered electrons in the SEM. J. Bone Miner. Res. 2(1), 13–22
(1987)
127. Rey, C., Collins, B., Goehl, T., Dickson, I.R., Glimcher, M.J.: The carbonate environment in
bone mineral: a resolution-enhanced fourier transform infrared spectroscopy study. Calcif.
Tissue Int. 45(3), 157–164 (1989)
128. Rey, C., Kim, H.M., Gerstenfeld, L., Glimcher, M.J.: Structural and chemical characteristics
and maturation of the calcium-phosphate crystals formed during the calcification of the
organic matrix synthesized by chicken osteoblasts in cell culture. J. Bone Miner. Res.
10(10), 1577–1588 (1995)
129. Rey, C., Renugopalakrishnan, V., Collins, B., Glimcher, M.J.: Fourier transform infrared
spectroscopic study of the carbonate ions in bone mineral during aging. Calcif. Tissue Int.
49(4), 251–258 (1991)
130. Riggs, B.L., Baylink, D.J., Kleerekoper, M., Lane, J.M., Melton 3rd, L.J., Meunier, P.J.:
Incidence of hip fractures in osteoporotic women treated with sodium fluoride. J. Bone
Miner. Res. 2(2), 123–126 (1987)
131. Riggs, C.M., Vaughan, L.C., Evans, G.P., Lanyon, L.E., Boyde, A.: Mechanical
implications of collagen fibre orientation in cortical bone of the equine radius. Anat.
Embryol. 187(3), 239–248 (1993)
132. Roschger, P., Fratzl, P., Eschberger, J., Klaushofer, K.: Validation of quantitative
backscattered electron imaging for the measurement of mineral density distribution in
human bone biopsies. Bone 23(4), 319–326 (1998)
133. Roschger, P., Fratzl, P., Klaushofer, K., Rodan, G.: Mineralization of cancellous bone after
alendronate and sodium fluoride treatment: a quantitative backscattered electron imaging
study on minipig ribs. Bone 20(5), 393–397 (1997)
134. Roschger, P., Gupta, H.S., Berzlanovich, A., Ittner, G., Dempster, D.W., Fratzl, P.,
Cosman, F., Parisien, M., Lindsay, R., Nieves, J.W., Klaushofer, K.: Constant
mineralization density distribution in cancellous human bone. Bone 32(3), 316–323 (2003)
135. Roschger, P., Rinnerthaler, S., Yates, J., Rodan, G.A., Fratzl, P., Klaushofer, K.:
Alendronate increases degree and uniformity of mineralization in cancellous bone and
decreases the porosity in cortical bone of osteoporotic women. Bone 29(2), 185–191 (2001)
136. Ross, P.D.: Risk factors for osteoporotic fracture. Endocrinol. Metab. Clin. North Am.
27(2), 289–301 (1998)
137. Roux, J.P., Wegrzyn, J., Arlot, M.E., Guyen, O., Delmas, P.D., Chapurlat, R., Bouxsein, M.L.:
Contribution of trabecular and cortical components to biomechanical behavior of human
vertebrae: an ex vivo study. J. Bone Miner. Res. 25(2), 356–361 (2010). doi:10.1359/jbmr.
090803
138. Sato, M., Westmore, M., Clendenon, J., Smith, S., Hannum, B., Zeng, G.Q., Brommage, R.,
Turner, C.H.: Three-dimensional modeling of the effects of parathyroid hormone on bone
distribution in lumbar vertebrae of ovariectomized cynomolgus macaques. Osteoporos. Int.
11(10), 871–880 (2000)
139. Schuit, S.C.E., van der Klift, M., Weel, A., de Laet, C., Burger, H., Seeman, E., Hofman, A.,
Uitterlinden, A.G., van Leeuwen, J., Pols, H.A.P.: Fracture incidence and association with
bone mineral density in elderly men and women: the Rotterdam study. Bone 34(1), 195–202
(2004)
140. Simmons Jr, E.D., Pritzker, K.P., Grynpas, M.D.: Age-related changes in the human femoral
cortex. J. Orthop. Res. 9(2), 155–167 (1991)
141. Smith, J.W.: The arrangement of collagen fibres in human secondary osteones. J. Bone Joint
Surg. Br. 42-B, 588–605 (1960)
142. Squillante, R.G., Williams, J.L.: Videodensitometry of osteons in females with femoral neck
fractures. Calcif. Tissue Int. 52(4), 273–277 (1993)
143. Stein, M.S., Feik, S.A., Thomas, C.D., Clement, J.G., Wark, J.D.: An automated analysis of
intracortical porosity in human femoral bone across age. J. Bone Miner. Res. 14(4),
624–632 (1999)
144. Stone, K.L., Seeley, D.G., Lui, L.Y., Cauley, J.A., Ensrud, K., Browner, W., Nevitt, M.C.,
Cummings, S.R.: BMD at multiple sites and risk of fracture of multiple types: long-term
results from the study of osteoporotic fractures. J. Bone Miner. Res. 18(11), 1947–1954
(2003)
145. Stout, S.D., Lueck, R.: Bone remodeling rates and skeletal maturation in three
archaeological skeletal populations. Am. J. Phys. Anthropol. 98(2), 161–171 (1995)
146. Takano, Y., Turner, C.H., Owan, I., Martin, R.B., Lau, S.T., Forwood, M.R., Burr, D.B.:
Elastic anisotropy and collagen orientation of osteonal bone are dependent on the
mechanical strain distribution. J. Orthop. Res. 17(1), 59–66 (1999)
147. Thompson, D.D.: Age changes in bone mineralization, cortical thickness, and haversian
canal area. Calcif. Tissue Int. 31(1), 5–11 (1980)
148. Thompson, D.D., Posner, A.S., Laughlin, W.S., Blumenthal, N.C.: Comparison of bone
apatite in osteoporotic and normal Eskimos. Calcif. Tissue Int. 35(3), 392–393 (1983)
149. Tong, W., Glimcher, M.J., Katz, J.L., Kuhn, L., Eppell, S.J.: Size and shape of mineralites
in young bovine bone measured by atomic force microscopy. Calcif. Tissue Int. 72(5),
592–598 (2003)
150. Vajda, E.G., Bloebaum, R.D.: Age-related hypermineralization in the female proximal
human femur. Anat. Rec. 255(2), 202–211 (1999)
151. Vincentelli, R., Evans, F.G.: Relations among mechanical properties, collagen fibers, and
calcification in adult human cortical bone. J. Biomech. 4(3), 193–201 (1971)
152. Von Ebner, V.: Ueber den feineren Bau der Knochensubstanz. Sber Akad Wiss Wien III 72,
49–138 (1875)
153. Wagermaier, W., Gupta, H.S., Gourrier, A., Burghammer, M., Roschger, P., Fratzl, P.:
Spiral twisting of fiber orientation inside bone lamellae. Biointerphases 1(1), 1–5 (2006)
154. Walker, R.A., Lovejoy, C.O., Meindl, R.S.: Histomorphological and geometric-properties
of human femoral cortex in individuals over 50––implications for histomorphological
determination of age-at-death. Am. J. Hum. Biol. 6(5), 659–667 (1994)
155. Wang, X.D., Li, X.O., Shen, X.M., Agrawal, C.M.: Age-related changes of noncalcified
collagen in human cortical bone. Ann. Biomed. Eng. 31(11), 1365–1371 (2003)
156. Wopenka, B., Pasteris, J.D.: A mineralogical perspective on the apatite in bone. Mater. Sci.
Eng. C-Biomimetic Supramol. Syst. 25(2), 131–143 (2005)
157. Wu, Y., Bergot, C., Jolivet, E., Zhou, L.Q., Laredo, J.D., Bousson, V.: Cortical bone
mineralization differences between hip-fractured females and controls. A Microradiogr.
Study Bone 45(2), 207–212 (2009)
158. Yeni, Y.N., Brown, C.U., Norman, T.L.: Influence of bone composition and apparent
density on fracture toughness of the human femur and tibia. Bone 22(1), 79–84 (1998)
159. Yerramshetty, J.S., Akkus, O.: The associations between mineral crystallinity and the
mechanical properties of human cortical bone. Bone 42(3), 476–482 (2008)
160. Yerramshetty, J.S., Lind, C., Akkus, O.: The compositional and physicochemical
homogeneity of male femoral cortex increases after the sixth decade. Bone 39(6),
1236–1243 (2006)
161. Zioupos, P.: Ageing human bone: factors affecting its biomechanical properties and the role
of collagen. J. Biomater. Appl. 15(3), 187–229 (2001)
Factor of Risk for Fracture
Dennis E. Anderson and Mary L. Bouxsein
Abstract In considering the risk of fracture, both the loading applied to a bone
and strength of the bone are of importance. A conceptually simple approach for
considering both loading and strength is the factor of risk, U, which is the ratio of
applied load to failure load for a particular loading scenario. Theoretically a
fracture will occur if U C 1. The factor of risk may provide a better measure for
risk of fracture than current clinical measures such as bone mineral density.
However, the challenges of accurately determining both applied load and failure
load are significant. A number of studies have examined factor of risk for hip,
vertebral and distal forearm fractures. At all three locations, factor of risk has been
found to increase with age, and to be associated with incident or prevalent frac-
tures. While some studies show promising results, the factor of risk has not been
consistently better than bone mineral density alone in predicting the risk of frac-
ture. However, it should be noted that the approaches used to estimate applied load
and failure load in most studies have been relatively simple. Furthermore, only a
few loading conditions have been investigated, primarily fall impact to the side for
the hip, forward flexion/lifting for the vertebral body and forward fall onto the
hand for the distal forearm. Thus, in spite of its limitations and challenges, factor
of risk may still provide significant insight into the etiology of osteoporotic
fractures, especially as methods for determining bone loading and strength
improve.
D. E. Anderson M. L. Bouxsein (&)

Department of Orthopedic Surgery, Center for Advanced Orthopaedic Studies,
Beth Israel Deaconess Medical Center, Harvard Medical School,
330 Brookline Ave, Boston, MA 02215, USA
e-mail: mbouxsei@caregroup.org

DOI: 10.1007/8415_2011_110
Published Online: 11 November 2011
134 D. E. Anderson and M. L. Bouxsein
1 Introduction to the Factor of Risk
From the simplest mechanical perspective, a fracture occurs when an applied load
on a bone exceeds its strength, or failure load. Thus, both loading and strength
must be considered when examining the risk of fractures. The loading experienced
by bones in vivo varies considerably depending on the specific activity. For
example the forces on the proximal femur during a fall impact will naturally be
greater than those during quiet standing. The strength of bones also varies
considerably, both between individuals and within an individual throughout life.
Bone strength depends in general on the geometry and material properties of the
bone [3, 68]. Of particular clinical relevance is bone mineral density (BMD),
which is the measure most commonly used to diagnose osteoporosis. Areal BMD
(aBMD) is defined as bone mineral content (BMC) divided by projected bone area,
typically measured using dual-energy X-ray absorptiometry (DXA). Because
BMD incorporates information about bone density and size, it provides a useful,
albeit imperfect, indirect measure of bone strength. aBMD accounts for about 50 to
80% of the variability in whole bone strength [4, 8, 37, 38, 48].
The importance of the interaction between skeletal loading and bone strength
has been demonstrated in several retrospective case-control studies. Nevitt and
Cummings [52] studied elderly women who fell and suffered a hip fracture
(n = 130), women who fell and suffered a wrist fracture (n = 294) and women
who fell and did not fracture (n = 467). They reported that among women who
fell on or near their hip, those who fell sideways or straight down were at fourfold
increased risk for hip fracture compared to those who fell in other directions,
whereas those who fell backward were less likely to suffer a hip fracture (odds
ratio1 = 0.2). Those who fell forward were more likely to suffer a wrist fracture.
Among those who fell either on their hand or hip, there was a twofold higher risk
of fracture for every standard deviation decrease in BMD. In another study,
Greenspan et al. [24] compared community-dwelling elderly individuals who fell
and suffered a hip fracture (n = 72) to those who fell and did not fracture
(n = 77). They found that low hip BMD, low body mass index and characteristics
related to the fall itself were independent risk factors for hip fracture (Table 1).
Taken together, these studies confirm an important interaction between bone
strength (as reflected by BMD), skeletal loading, and fracture risk.
Insight into the relative contributions of skeletal strength and skeletal loading to
the etiology of fracture may be gained by evaluating the ratio of load applied to the
bone (applied load) to strength of bone (failure load). This comparison of applied
load versus failure load gives an estimate of how ‘‘safe’’ the structure is from
failure, and is termed the ‘‘factor of risk’’, U [3, 26, 34, 68]:
U = Applied load/Failure load
1
odds ratio (OR) is defined as the odds of an event occurring in one group, divided by the odds
of the it occurring in another group.
Factor of Risk for Fracture 135
Table 1 Multiple logistic regression analysis of factors associated with hip fracture among 149
community-dwelling men and women who fell (data from [24])
Factor Adjusted odds ratio 95% Confidence interval p-value
Fall to the sidea 5.7 2.3–14 \0.001
Femoral neck aBMD (g/cm2)b 2.7 1.6–4.6 \0.001
Potential energy of the fall (J)c 2.8 1.5–5.2 \0.001
Body mass index (kg/m2)b 2.2 1.2–3.8 0.003
a
OR versus any other direction of fall
b
OR calculated for a decrease of 1 SD
c
OR calculated for an increase of 1 SD
Theoretically, when the factor of risk is less than one the forces applied to the
bone are insufficient to fracture it, and the bone will not fracture. However, when
the factor of risk is greater than or equal to one (i.e., applied load equals or exceeds
failure load), fracture is predicted to occur. A high factor of risk can occur either
when the bone is weak and its load bearing capacity is compromised, or when very
high loads, such as those resulting from a fall or other trauma, are applied to the
bone. In elderly individuals, it is likely that the coupling of a weak bone with an
increased incidence of traumatic loading leads to the dramatic rise in fracture
incidence with age [5, 63].
Determining the factor of risk during an event such as a fall impact requires
determining applied loading and failure loading specific to that event (Fig. 1).
While the factor of risk is a simple concept, in practice obtaining accurate estimates
of bone strength and skeletal loading are both significant challenges. Additionally,
there are multiple ways for strength and loading to be expressed. The components
of factor of risk may be in terms of force or moment, in which case factor of risk
represents the risk of the whole bone to fail under an applied force or moment,
respectively. Alternatively, they may be in terms of stress, that is applied stress and
material strength. Duan and coworkers have used a stress-based factor of risk,
which was termed the ‘‘Fracture Risk Index’’, in studies of vertebral fracture risk
[17–19]. In any case, the units of applied load and failure load must be equivalent as
the factor of risk itself is unitless. Risk of fracture may also be evaluated using the
inverse of the factor of risk, which is called the factor of safety. Factor of safety has
occasionally been used in examining the risk of fracture in bone [23, 35].
1.1 Determining the Forces Applied to Bone in Vivo
Determining the numerator of factor of risk, loading applied to bone, is difficult

because of the complexity of in vivo bone loading and the large number of pos-
sible loading conditions. Bone loading in vivo includes numerous forces applied
simultaneously via interactions with muscles, tendons, ligaments, other bones and
external forces. However, to date, the approach to determining applied loading in
factor of risk studies has been relatively simple, based primarily on basic bio-
mechanical models or empirical equations based on anthropometric data such as
Select Activity or Event of Interest

(e.g. fall impact, lifting a heavy load)
Determine Applied Load: Determine Failure Load
Depends on: Depends on:

• Specific Activity • Bone Mineral Density
• External Forces • Bone Size & Shape
• Anthropometry (e.g. • Microarchitecture
height and weight) • Damage Accumulation
• Muscle Forces • Degree of Mineralization
• Soft Tissue Padding or • Loading Direction
Attenuation • Loading Mode (e.g.
bending, torsion,
compression)
• Loading Rate
If Φ < 1 ,
Calculate Factor of Risk: No Fracture
Applied Load
Φ=
Failure Load If Φ ≥1,
Fracture
Fig. 1 Schematic of the process for determining the factor of risk. The applied load and failure
load should both be representative of the same activity or event, giving the factor of risk specific
to that activity or event. Determinants of applied load and failure load are noted. Theoretically, if
accurately determined, the factor of risk will indicate whether fracture will occur
height and weight. In addition, relatively few loading conditions have been studied
using factor of risk. For example, studies of wrist fractures have used only a
forward fall with loading on the outstretched hand; studies of vertebral fractures
have examined standing and forward flexion with and without lifting a weight; and
the majority of studies of hip fractures have examined a single sideways fall
condition. There is a need to examine a greater variety of loading conditions to
study the full spectrum of activities and situations that may lead to fractures.
1.2 Estimating Whole Bone Strength in Vivo
The denominator of factor of risk, bone failure load or strength, may be determined
in several ways. Generally, human cadaveric specimens are measured with one or
more non-invasive imaging modalities (e.g., quantitative computed tomography

(QCT), or DXA) and then mechanically tested to failure in loading conditions
designed to mimic those that cause fractures. Then, using those data, whole bone
strength is estimated in vivo using regression equations between whole bone
strength and aBMD, volumetric bone mineral density (vBMD) and/or other mass
or geometric features derived from the imaging modalities. Whole bone strength
may also be estimated from finite element models based on QCT or high-reso-
lution peripheral QCT (HR-pQCT). These approaches are described in more detail
elsewhere, but are generally better predictors of bone strength than bone density
measurements alone [10, 12, 33, 60, 75].
It is important to remember that the strength of a bone depends greatly on the
loading conditions, that is the strength for one activity may not be the same as for
another. For example, a proximal femur withstands much higher force magnitudes
when tested in a single-leg stance configuration than it does when tested in a
configuration designed to simulate a sideways fall [35, 38, 39]. Even different
directions of sideways falls can significantly impact femoral strength [59]. In
general, bone is weaker in shear or tension conditions than in compression
[61, 72]. Furthermore, yield strength of bone decreases with higher strain rates
[25], which may be of importance for fracture under traumatic conditions such as
fall impact loading. Thus, characterizing the activities that cause fractures and
reproducing those loading conditions for the estimates of bone strength is critical,
though challenging.
2 Examinations of Factor of Risk
Once estimates of both loading and bone strength are made, the factor of risk can
be calculated. In recent years, a number of retrospective studies and a few pro-
spective studies have examined the factor of risk for fractures of the hip, vertebral
body and wrist. These are all common fractures in osteoporotic adults, together
representing an estimated 60% of osteoporotic fractures [7]. This section critically
reviews this literature, which represents early attempts to incorporate both loading
and strength when examining the risk of fractures. We pay particular attention to
(1) the study design, as prospective studies are the gold standard for examining
predictors of fracture risk, and (2) whether current implementations of the factor of
risk perform better than BMD in predicting fracture risk.
2.1 Studies of the Factor of Risk for Hip Fracture
Hip fractures are common injuries in older adults and their incidence is increasing;
they are costly and associated with high rates of morbidity and mortality [7, 41, 57].
More than 90% of hip fractures are associated with a fall [41], but only about 1%
of falls in the elderly are associated with a hip fracture [14, 28]. This indicates that
the mechanics of the fall can have a major effect on the risk of fracture. For
example, falling to the side and impacting the hip or side of the leg increases the risk
of hip fracture approximately 20-fold relative to falling in any other direction [27].
A majority of hip fractures occur in individuals not classified as osteoporotic based
on BMD [67, 69–71, 76]. Thus, a factor of risk approach may be able to distinguish
the risk of hip fractures better than BMD.
To date, all studies of factor of risk for hip fracture have used a fall on the hip to
the side from standing height as the loading condition of interest. Loading has been
estimated based on laboratory studies of fall dynamics and impact mechanics
[65, 66, 73, 74]. Thus, forces on the hip during a sideways fall are estimated using
subject-specific body height and weight, along with stiffness and damping
constants derived from these laboratory experiments [6, 34, 50, 54, 56, 63, 64, 77].
The loading attenuation due to variation in the thickness of trochanteric soft tissues
was examined in two studies [6, 54], whereas one study employed a uniform
trochanteric soft tissue thickness [56].
The denominator of the factor of risk, or load-bearing capacity of the proximal
femur, has been determined from linear regressions between BMD and femoral
failure loads in a fall configuration in several studies [6, 50, 54, 77], whereas one
study used engineering beam-theory to predict femoral strength [63]. Two recent
studies estimated failure load using QCT-based finite element analysis [34, 56],
and another measured failure load directly by mechanical testing of cadaveric
specimens [64].
2.1.1 Age- and Osteoporosis-Related Changes in the Factor

of Risk for Hip Fracture
A study of 788 (548 females, 240 males) healthy individuals aged 21–77 years in
Taiwan examined the change in factor of risk with age [77]. Factor of risk
increased with age in younger men (\50), but was relatively constant with age in
older men. However, in women factor of risk increased with age throughout life,
with the average factor of risk approaching one by age 80.
Riggs et al. [63] reported age- and sex-specific differences in the load to strength
ratio at the femoral neck in 700 men and women aged 21–97. They found that the
load to strength ratios for bending and axial loading were only marginally higher
(i.e, worse) in young women versus young men, but that the ratios increased
(i.e, worsened) nearly twofold more over life in women (+40–62%) compared
to men (+22–36%). This increase in the load to strength ratio was attributable to
greater declines in cortical and trabecular vBMD at the femoral neck in women than
men [62] as there was negligible age-related change in the load applied to the hip.
While this pattern explains some of the observed increase in hip fracture risk in
women, it did not fully explain the fourfold increase in hip fracture risk with age,
suggesting that additional factors not represented by the load to strength ratio, such
as greater incidence of falls, contribute to the increased risk of hip fracture with age.
Factor of risk analysis shows that both osteopenic and osteoporotic individuals
are likely to suffer a hip fracture in a fall. Based on analysis of 166 post-meno-
pausal women using QCT-based finite element analysis to estimate femoral
strength, 87% of osteoporotic (i.e femoral BMD t-score B -2.5) and 84% of
osteopenic (femoral BMD t-score between -1 and -2.5) women were likely to
have a hip fracture in a fall to the side, whereas only 35% of women with normal
BMD (femoral BMD t-score [ -1) were predicted to fracture during fall to the
side [34]. In addition, factor of risk was found to be only weakly correlated
(r2 = 0.14) with total hip aBMD. In a somewhat different approach, Roberts et al.
[64] examined factor of risk in 73 cadaveric femurs, using the directly measured
strength value for a sideways fall configuration, and estimating fall loads from
subject height and weight. Nearly all osteoporotic femurs (femoral neck
t-score \ -2.5) and half of non-osteoporotic femurs, had a factor of risk [1.0,
indicating that they would be at high risk for a fracture in a sideways fall, whereas
only one in six femurs with femoral BMD t-score [ -1 had a factor of risk
value C1.0.
2.1.2 Association Between the Factor of Risk and Hip Fracture
Several studies have shown that factor of risk for hip fracture is increased in both
men and women with incident hip fractures. In an early retrospective study, Myers
et al. [50] reported a strong association between the factor of risk and hip fracture
in 231 elderly fallers, 98 cases with hip fracture and 138 controls without fracture.
Yang et al. [77] found 26 women with hip fractures to have higher factor of risk
than 85 healthy age-matched controls. In a prospective, nested case-control study,
21 postmenopausal women with incident hip fracture had reduced trochanteric soft
tissue thickness, reduced femoral aBMD and increased factor of risk compared to
42 age-matched controls [6]. The force applied to the femur was calculated both
with and without attenuation due to trochanteric soft tissue. Without soft tissue
attenuation the factor of risk was greater than one in both cases and controls, while
with soft tissue attenuation the factor of risk approached one in cases and was less
than one in controls (0.92 ± 0.44 and 0.62 ± 0.50, respectively). In this study, the
association with hip fracture was similar for femoral BMD (OR = 2.1, CI 1.2–3.5)
and the factor of risk (OR = 1.85, CI 0.96–3.6).
Orwoll et al. [56] studied the factor-of-risk for hip fracture in a prospective
study of community-dwelling men over age 65 with an average of 5.6 years of
followup. In a nested case-cohort analysis of 40 men who suffered a hip fracture
and 210 who did not, femoral strength was determined from baseline QCT scans
using finite element analysis and fall loads estimated using a fixed trochanteric soft
tissue thickness of 25 mm. Factor of risk was higher in fracture cases than non-
cases (1.13 ± 0.41 vs. 0.75 ± 0.24; p \ 0.01). Cox proportional hazards regres-
sion indicated that low femoral aBMD (hazard ratio, HR = 4.4, 95% CI 2.1–9.1),
low femoral strength (HR = 6.5; 95% CI 2.3–18.3) and high factor-of risk
(HR = 4.3, 95% CI 2.5–7.4) were significant risk factors for hip fracture,
adjusting for age and BMI. Importantly, after additionally adjusting for femoral
aBMD, the factor of risk remained a significant risk factor for hip fracture
(HR = 3.1, 95% CI 1.6–6.1) while femoral strength did not (HR = 2.7, 95% CI
0.5–14.6).
2.2 Studies of the Factor of Risk for Vertebral Fractures
Vertebral fractures are the most common osteoporotic fracture, accounting for
about 27% of all osteoporotic fractures [7]. Although only a minority of radio-
graphically evident vertebral deformities come to clinical attention [11, 16, 53],
they are associated with significant morbidity and are strong predictors of future
fracture risk [15, 36, 53]. Unlike the clear association between hip fractures and
falls, the events and resultant spinal loading associated with vertebral fractures
remain unclear. In a hospital-based study, nearly 50% of acute, symptomatic
vertebral fractures in individuals over age 60 were associated with a fall, whereas
20% were associated with ‘controlled’ activities, such as bending, lifting, and
reaching [51]. In a prospective study of older men, approximately 57% of incident
clinically evident vertebral fractures were associated with a fall, but 21% occurred
in unknown circumstances [22]. In a retrospective study of circumstances asso-
ciated with clinically diagnosed vertebral fractures, a specific loading event was
reported for only about 50% of the total fractures [11]. Of this 50%, 10% of
fractures were associated with ‘‘lifting a heavy object’’, whereas nearly 40% were
associated with falling. Thus, while it appears that many vertebral fractures are due
to falls, many occur in unknown conditions. Since loads are applied to the spine
during every activity of daily living, it is crucial to distinguish which activities
are most strongly associated with increased risk of vertebral fractures.
Studies of factor of risk for vertebral fractures have primarily examined forward
flexion and/or forward flexion while lifting a weight [5, 17–19, 44, 49]. Most of
these studies have estimated vertebral loads using simple biomechanical models
that include only a single spinal extensor muscle and have focused on the L3
vertebra. The exception is the work of Myers and Wilson [49], which examined the
L2 vertebra and used an optimization approach to estimate muscle forces and
vertebral compression. Such an approach may be more appropriate in estimating
vertebral loading, as it includes multiple trunk muscles and allows for other
activities to be modeled, such as twisting or asymmetric lifting, such as carrying a
suitcase. No studies have examined factor of risk for vertebral fracture during a
fall, likely due to the difficulty in estimating vertebral loading during a fall.
The majority of studies examining vertebral factor of risk have estimated ver-
tebral strength using empirical relationships with BMD. Myers and Wilson used a
linear relationship between strength and aBMD [49], while Duan and coworkers
used a power law relationship between strength and vBMD [17–19]. Bouxsein et al.
[5] determined strength from the vertebral body elastic modulus and cross-sectional
area, where elastic modulus was estimated from a linear relationship with vBMD.
(The product of elastic modulus and area is sometimes referred to as ‘axial

rigidity’.) Unlike other studies to date, Melton and coworkers have determined
vertebral body strength using QCT-based finite element models [44, 47].
Using a modeling approach, Myers and Wilson [49] examined the how factor of
risk for fracture for a variety of activities of daily living in the L2 vertebra varied
with vertebral aBMD. Their analyses predicted that individuals with extremely low
BMD (e.g. 0.3 g/cm2) may be at risk for vertebral fracture during simple activities
such as tying one’s shoes or opening a window. Individuals with low BMD (e.g. up
to 0.6 g/cm2) may be at risk for vertebral fracture during bending and lifting tasks
such as picking up a toddler. This shows that for osteoporotic individuals, vertebral
fractures may occur in non-traumatic controlled activities.
2.2.1 Age-, Sex- and Ethnicity-Related Differences in the Factor

of Risk for Vertebral Fracture
Application of the factor-of-risk for vertebral fracture in a population-based

sample of 697 women and men aged 20–97 years found interactions between age
and sex [5]. Factor of risk during forward bending was 0.42 ± 0.08 in young men,
0.36 ± 0.08 in young women, 0.50 ± 0.14 in older men and 0.59 ± 0.15 in older
women. Thus, while the factor of risk increased with age in both men and women
(p \ 0.01), the increase was more than three times larger in women than men
(p \ 0.01). A similar interaction was seen for factor of risk in a lifting task. This
was in part due to the fact that while both sexes exhibited a marked decline in
lumbar vertebral compressive strength with age (p \ 0.001), the decline was
greater in women than men (-49 vs.-31%, p \ 0.001). For bending forwards and
lifting, the factor of risk in individuals age 50 and over exceeded one (U [ 1) in
30% of women and 12% of men, similar to reported prevalence of vertebral
fracture. Similarly, an examination of 1,013 healthy volunteers [17] found the
factor of risk for vertebral fracture when bending forward 90° to increase signif-
icantly with age in both men and women (p B 0.01), but more so in women. In
individuals age 60 and over, the factor of risk exceeded one in 26% of women and
9% of men.
In a cross-sectional study of 1,868 healthy men and women aged 18–91,
factor of risk was compared between individuals of Chinese and Caucasian race
[18]. In young adults, Chinese had a lower factor of risk than Caucasians in both
men and women (p \ 0.01); in elderly adults, Chinese men had lower factor of
risk than Caucasian men, but Chinese women had higher factor of risk than
Caucasian women (p \ 0.01). However, only 5% of Chinese and 6% of Cau-
casian elderly men had U [ 1, whereas 25% of Chinese and 29% of Caucasian
elderly women had U [ 1. The similar proportions of elderly Chinese and
Caucasian adults with high factor of risk is consistent with the fact that no
differences in incidence of vertebral fractures have been found between Chinese
and Caucasian populations.
2.2.2 Association Between the Factor of Risk and Vertebral Fracture
Factor of risk is increased in individuals with vertebral fracture. In a cross-sectional

study of 81 patients with vertebral fractures [17] the mean factor of risk, using
forward flexion and vertebral strength estimated from spine BMD, was greater than
or equal to one for patients with vertebral fractures (men: 1.03; women: 1.35;
p \ 0.05) and was greater than the factor of risk for age-matched controls without
fracture (men: 0.76; women: 1.06; p \ 0.01). Moreover, in a study comparing 40
post-menopausal women with prevalent vertebral fracture to 40 without, the esti-
mated factor of risk for vertebral fracture (simulating 90o forward flexion and using
QCT-based finite element analysis (FEA) to estimate vertebral strength) was
significantly higher in subjects with prevalent vertebral fractures than in those with
no fractures. The OR’s for three different predictors of fracture were: factor of
risk = 3.2 (95% CI 1.4-7.5), FEA vertebral strength = 2.2 (95% CI 1.2-4.3) and
integral vBMD = 2.2 (95% CI 1.1-4.3) [44].
A more recent study examined the association between vBMD, vertebral
strength and the factor of risk in postmenopausal women with mild (n = 141) vs
moderate/severe (n = 52) prevalent vertebral deformities [47]. Vertebral strength
was estimated from QCT-based FEA, and activities included upright standing, plus
forward flexion with and without weights. Women with mild vertebral deformities
had lower vertebral strength and increased factor of risk compared to women with
no fractures, whereas women with moderate and severe fractures had even lower
vertebral strength and higher factor of risk values for several activities. Associa-
tions between vertebral fracture and vBMD, vertebral strength and factor of risk
were higher in women with moderate/severe fractures (OR = 2.9-3.5) than in
women with mild fractures (OR = 1.4-1.5), but were similar amongst different
predictors of fracture.
The use of factor of risk as a predictor for vertebral fracture was compared to
use of aBMD in both a cross-sectional case-control study with 89 postmenopausal
women with fractures and 306 controls, as well as a 10-year prospective study in
which 30 women with incident fractures were compared to 150 women without
[19]. In the case-control study, factor of risk was significantly associated with
prevalent fracture (OR 2.06; 95% CI 1.55-2.73), but was less sensitive and
specific than aBMD in distinguishing cases and controls. In the prospective study,
factor of risk was not predictive of fracture (HR 1.20; 95% CI 0.9-1.7), but aBMD
was (HR 2.4; 95% CI 1.5-3.8). This study concluded that factor of risk is not
better than aBMD in predicting the risk of fracture.
2.3 Studies of the Factor of Risk for Distal Forearm Fracture
Fractures of the distal forearm or wrist are the second most common osteoporotic
fracture after vertebral fractures, accounting for about 19% of fractures [7].
Excluding severe trauma, fractures of the distal forearm in adults over 50 are
almost always caused by a fall onto the hand [58]. Distal forearm fractures are
predictive of increased risk of subsequent hip or vertebral fractures in both men
and women [13].
Studies of factor of risk for distal forearm fracture have used a forward fall onto
the hand as the loading condition of interest. In all of these studies, loading has
been estimated as a damping coefficient times impact velocity [32, 45, 46, 48]. The
damping coefficient was experimentally determined, and impact velocity estimated
from fall height [9]. The strength of the distal forearm in several recent studies of
factor for risk has been based on micro-finite element models [32, 45, 46]. This
approach is excellent for examinations of the distal forearm, as micro-finite ele-
ment models can be developed from HR-pQCT scans. An earlier study of distal
forearm factor of risk used mechanical testing of cadaveric radii to determine
distal forearm strength [48].
2.3.1 Age-, Sex- and Osteoporosis-Related Differences in the Factor

of Risk for Forearm Fractures
In a mechanical testing study, fracture strength was measured in cadaveric spec-

imens from nine female and 12 male donors [48] and factor of risk estimated for a
forward fall onto the hand. Mean factor of risk was 1.04 in females versus 0.79 in
males (p \ 0.01). Thus, this small study indicates that women are at greater risk of
fracturing the distal radius than men. The factor of risk was negatively correlated
with distal radius T-score (R2 = 0.73), with a T-score below -1.5 indicating a
high risk of fracture.
Application of the factor-of-risk for wrist fracture in a population-based sample
of 700 women and men aged 20–97 years [63] showed that men had a higher
predicted forearm strength at all ages, largely due to their greater bone size.
In young adults, load to strength ratios at the distal radius were lower (better) in
men than in women, and increased (worsened) over life significantly more in
women than in men. These patterns are consistent with the higher rate of forearm
fractures in older women than men, and lack of marked age-related increase in wrist
fractures incidence in men.
Kazakia et al. [32] examined the variation of densitometric, geometric,
microstructural and biomechanical parameters with aBMD in one cohort of 58
post-menopausal women and another cohort of 142 men and women ages 20–78.
Factor of risk for distal forearm fracture was negatively correlated with aBMD
(R2 = 0.80, p \ 0.0001), and was higher in osteoporotic individuals (p \ 0.01).
2.3.2 Association Between the Factor of Risk and Forearm Fractures
Factor of risk is increased in individuals with prevalent and incident forearm

fractures. In a cross-sectional, case-control study, 18 post-menopausal women
with a distal forearm fracture were compared with 18 age-matched controls [45].
Factor of risk was 24% higher in cases as compared to controls (p \ 0.01).

However, vBMD (OR = 4.2; 95% CI 1.4–12), cortical thickness (OR = 4.0; 95%
CI 1.4–11), and bone axial rigidity (OR 3.8; 95% CI 1.4–10) were all similar
predictors of fracture compared to factor of risk (OR = 3.0; 95% CI 1.2–7.5). In
another cross-sectional case–control study, 100 post-menopausal women with
prevalent distal forearm fracture were compared to 105 without [46]. Factor of risk
was 15% higher in those with prevalent fractures (OR 1.9; 95% CI 1.4–2.6).
However, factor of risk did not offer an improvement over aBMD (OR 2.0; 95% CI
1.4–2.8) in terms of clinical assessment of fracture risk. In addition, the mean
factor of risk among controls was greater than one, indicating possible under-
estimation of bone strength and/or overestimation of fall loads. Another inter-
pretation would be that most post-menopausal women would be at high risk for
fracture when falling on the hand. Finally, factor of risk was reported for a cross-
sectional case-control study including 33 women with incident wrist fractures
and 33 controls matched for age, height, weight and age at menopause taken from
a prospective cohort examining determinants of bone loss [2]. Factor of risk was
about 16% higher in cases than in controls (1.08 ± 0.16 vs. 0.93 ± 0.19,
p \ 0.001).
3 Discussion
The factor of risk is a conceptually simple approach for estimating the risk of
fracture that considers both the strength of bone and loading applied to bone, and
thus offers a theoretical advantage compared to measures based only on bone
strength. However, studies examining factor of risk to date have had mixed results;
there has not been compelling evidence that factor of risk provides a major
improvement in predicting the risk of a fracture over common measures such as
BMD. Inaccuracies and uncertainties in estimates of loading and/or strength could
easily reduce the usefulness of the approach, as factor of risk does not account for
variability or uncertainty in either strength or loading estimates. An alternative,
albeit more complex, approach that can account for such variability is the use of
injury risk functions [29].
While its simplicity makes factor of risk an attractive approach, this may belie
the challenges associated with accurately determining loading and strength of
bone. Examinations of factor of risk to date have primarily used simple estimates
of loading. However, in vivo bone loading is complex, and may not always be
easily represented by a single applied load. For example, muscle contraction
during fall descent may help in absorbing energy, but muscle contraction during
impact can increase the impact load by 25–100% [28]. Muscle action may also act
to protect bone, as muscular contraction significantly increases the load and energy
required to fracture the tibia in rats [55]. Thus, muscle action may either increase
or reduce the loading applied to bone, and thereby affect the factor of risk.
The determination of muscle forces in vivo could improve the estimation of

bone loading in factor of risk studies. A common approach for estimating muscle
forces is the use of a biomechanical model that represents the musculoskeletal
anatomy and determines muscle forces using an optimization algorithm. Only one
study to date has used this approach in estimating loading for factor of risk [49].
Such models have been used to estimate forces on the proximal femur during gait
[30] and vertebral bodies during lifting tasks [1]. However, it is unknown if muscle
forces can be accurately determined during an event such as a fall using such
models. Nonetheless, future examinations of factor of risk may benefit from
improved estimation of muscle forces, and therefore skeletal loading, using more
complex biomechanical models.
Studies of factor of risk have not examined all relevant loading conditions,
particularly in the case of the spine. Vertebral factor of risk during a fall has not been
examined even though up to half of acute vertebral fractures occur during a fall
[11, 22, 51]. Moreover, in a finite element study, Matsumoto et al. [42] showed that
fracture force for L2 was significantly lower in a forward bending configuration than
in uniaxial compression of the vertebral body. Thus, while uniaxial loading has been
used in factor of risk studies of vertebral bodies, it may not be the most relevant
loading configuration for vertebral fracture. Finally, studies of factor of risk have
primarily focused on the L3 vertebral level, while clinically most vertebral fractures
occur at mid-thoracic (T7–T8) and thoraco-lumbar (T11–L1) locations [20, 22, 31,
40, 43]. Studies examining the variation in vertebral strength, loading and factor of
risk at different spinal locations may provide additional insights.
Even in the distal forearm, where fractures are almost invariably caused by a
fall onto the hand and muscle forces may not play a significant role, there are other
possible loading conditions. All studies of factor of risk for forearm fracture have
used the data of Chiu and Robinovitch [9], based on a forward fall, to estimate
loading. However, only about 40–45% of falls that cause a distal forearm fracture
are in the forward direction, with about 40–45% backward, and about 15% to the
side [46, 58]. Falls in these other directions could possibly produce different
loading conditions on the distal forearm than a forward fall.
Similar to the current limitations for estimates of skeletal loading, estimation of
bone strength for factor of risk in some studies has been based on relatively simple
approaches. Specifically, many studies have based strength on regression equations
relating strength to aBMD, which may not reflect all the aspects contributing to
whole bone strength. In comparison, a number of studies of factor of risk have
determined strength using state-of-the art QCT- or HR-pQCT-based finite element
models, including in the proximal femur [34, 56], vertebral body [44, 47] and
distal forearm [32, 45, 46]. This may be the best approach for determining in vivo
bone strength. Finite element models also have the advantage that strength can be
determined for multiple and complex loading conditions. However, strength
estimates from finite element analyses are greatly dependent on the imposed
boundary conditions, and future work could examine the effect of different
boundary conditions on the FEA-based strength estimates. As already noted ver-
tebral strength estimates are different when applying axial compression versus a
more realistic loading condition of axial compression plus forward bending.

In addition, studies of vertebral factor of risk using FEA have applied loads to the
vertebral bodies through PMMA-endplates, simulating cadaveric testing methods
[44, 47]. However, in vivo loads are applied through the intervertebral disc, and
recent work has shown that vertebral body endplates may experience high tensile
strains due to the Poisson expansion of the disc, placing them at high risk for
failure [21], whereas PMMA endplates would produce a very different strain
distribution.
Given accurate estimates of bone loading and strength, factor of risk will
indicate the risk of a fracture under a particular loading condition. However, this
does not account for the likelihood of that loading condition occurring. That is,
factor of risk indicates the risk of a hip fracture occurring in a fall, but not the risk
of falling. With improved loading and strength estimates, factor of risk may
improve identification of at-risk individuals. In addition, studies of factor of risk
may provide a way to identify activities that place individuals at high risk for
fractures. For example, identifying non-traumatic loading scenarios that have
increased factor of risk for vertebral fractures may aid in the prevention of oste-
oporotic fractures.
In conclusion, it is clear that both loading and strength are important in fracture
etiology. Thus, in spite of its limitations and challenges, factor of risk remains a
useful concept for examining osteoporotic fractures. Future work in biomechanics
will continue to improve estimates of loading and strength of bone, and thereby the
utility of the factor of risk for predicting fracture risk.
Acknowledgments We would like to acknowledge funding from the National Institutes of

Health: R01AR053986 and a postdoctoral fellowship from the Harvard Translational Research in
Aging Training Program (T32AG023480).
References
1. Arjmand, N., Shirazi-Adl, A.: Sensitivity of kinematics-based model predictions to

optimization criteria in static lifting tasks. Med. Eng. Phys. 28(6), 504–514 (2006)
2. Boutroy, S., Van Rietbergen, B., Sornay-Rendu, E., Munoz, F., Bouxsein, M.L., Delmas, P.D.:
Finite element analysis based on in vivo HR-pQCT images of the distal radius is associated with
wrist fracture in postmenopausal women. J. Bone Miner. Res. 23(3), 392–399 (2008).
doi:10.1359/jbmr.071108
3. Bouxsein, M.: Biomechanics of age-related fractures. In: Marcus, R., Feldman, D., Nelson, D.,
Rosen, C. (eds.) Osteoporosis, vol 1, vol. 3, pp. 601–616. Elsevier Academic Press, San Diego
(2007)
4. Bouxsein, M.L., Coan, B.S., Lee, S.C.: Prediction of the strength of the elderly proximal
femur by bone mineral density and quantitative ultrasound measurements of the heel and
tibia. Bone 25(1), 49–54 (1999). doi:S8756-3282(99)00093-9
5. Bouxsein, M.L., Melton 3rd, L.J., Riggs, B.L., Muller, J., Atkinson, E.J., Oberg, A.L.,
Robb, R.A., Camp, J.J., Rouleau, P.A., McCollough, C.H., Khosla, S.: Age- and sex-specific
differences in the factor of risk for vertebral fracture: a population-based study using QCT.
J. Bone Miner. Res. 21(9), 1475–1482 (2006)
6. Bouxsein, M.L., Szulc, P., Munoz, F., Thrall, E., Sornay-Rendu, E., Delmas, P.D.:
Contribution of trochanteric soft tissues to fall force estimates, the factor of risk, and
prediction of hip fracture risk. J. Bone Miner. Res. 22(6), 825–831 (2007). doi:10.1359/
jbmr.070309
7. Burge, R., Dawson-Hughes, B., Solomon, D.H., Wong, J.B., King, A., Tosteson, A.:
Incidence and economic burden of osteoporosis-related fractures in the United States,
2005–2025. J. Bone Miner. Res. 22(3), 465–475 (2007). doi:10.1359/jbmr.061113
8. Cheng, X.G., Lowet, G., Boonen, S., Nicholson, P.H., Brys, P., Nijs, J., Dequeker, J.:
Assessment of the strength of proximal femur in vitro: relationship to femoral bone mineral
density and femoral geometry. Bone 20(3), 213–218 (1997). doi:S8756328296003833
9. Chiu, J., Robinovitch, S.N.: Prediction of upper extremity impact forces during falls on the
outstretched hand. J. Biomech. 31(12), 1169–1176 (1998)
10. Cody, D.D., Gross, G.J., Hou, F.J., Spencer, H.J., Goldstein, S.A., Fyhrie, D.P.: Femoral
strength is better predicted by finite element models than QCT and DXA. J. Biomech. 32(10),
1013–1020 (1999). doi:S0021929099000998[pii]
11. Cooper, C., Atkinson, E.J., O’Fallon, W.M., Melton 3rd, L.J.: Incidence of clinically
diagnosed vertebral fractures: a population-based study in Rochester, Minnesota, 1985–1989.
J. Bone Miner. Res. 7(2), 221–227 (1992)
12. Crawford, R.P., Cann, C.E., Keaveny, T.M.: Finite element models predict in vitro vertebral
body compressive strength better than quantitative computed tomography. Bone 33(4),
744–750 (2003)
13. Cuddihy, M.T., Gabriel, S.E., Crowson, C.S., O’Fallon, W.M., Melton 3rd, L.J.: Forearm
fractures as predictors of subsequent osteoporotic fractures. Osteoporos. Int. 9(6), 469–475
(1999)
14. Cummings, S.R., Marcus, R., Palermo, L., Ensrud, K.E., Genant, H.K.: Does estimating
volumetric bone density of the femoral neck improve the prediction of hip fracture? A
prospective study. Study of Osteoporotic Fractures Research Group. J. Bone Miner. Res.
9(9):1429–1432 (1994)
15. Delmas, P.D., Genant, H.K., Crans, G.G., Stock, J.L., Wong, M., Siris, E., Adachi, J.D.:
Severity of prevalent vertebral fractures and the risk of subsequent vertebral and nonvertebral
fractures: results from the MORE trial. Bone 33(4), 522–532 (2003)
16. Delmas, P.D., van de Langerijt, L., Watts, N.B., Eastell, R., Genant, H., Grauer, A.,
Cahall, D.L.: Underdiagnosis of vertebral fractures is a worldwide problem: the IMPACT
study. J. Bone Miner. Res. 20(4), 557–563 (2005)
17. Duan, Y., Seeman, E., Turner, C.H.: The biomechanical basis of vertebral body fragility in
men and women. J. Bone Miner. Res. 16(12), 2276–2283 (2001). doi:10.1359/jbmr.2001.
16.12.2276
18. Duan, Y., Wang, X.F., Evans, A., Seeman, E.: Structural and biomechanical basis of racial
and sex differences in vertebral fragility in Chinese and Caucasians. Bone 36(6), 987–998
(2005). doi:10.1016/j.bone.2004.11.016
19. Duan, Y., Duboeuf, F., Munoz, F., Delmas, P.D., Seeman, E.: The fracture risk index and
bone mineral density as predictors of vertebral structural failure. Osteoporos. Int. 17(1),
54–60 (2006). doi:10.1007/s00198-005-1893-5
20. Eastell, R., Cedel, S.L., Wahner, H.W., Riggs, B.L., Melton 3rd, L.J.: Classification of
vertebral fractures. J. Bone Miner. Res. 6(3), 207–215 (1991)
21. Fields, A.J., Lee, G.L., Keaveny, T.M.: Mechanisms of initial endplate failure in the human
vertebral body. J. Biomech. 43(16), 3126–3131 (2010). doi:10.1016/j.jbiomech.2010.08.002
22. Freitas, S.S., Barrett-Connor, E., Ensrud, K.E., Fink, H.A., Bauer, D.C., Cawthon, P.M.,
Lambert, L.C., Orwoll, E.S.: Rate and circumstances of clinical vertebral fractures in older
men. Osteoporos. Int. 19(5), 615–623 (2008). doi:10.1007/s00198-007-0510-1
23. Fritz, J.M., Guan, Y., Wang, M., Smith, P.A., Harris, G.F.: A fracture risk assessment model
of the femur in children with osteogenesis imperfecta (OI) during gait. Med. Eng. Phys.
31(9), 1043–1048 (2009). doi:10.1016/j.medengphy.2009.06.010
24. Greenspan, S.L., Myers, E.R., Maitland, L.A., Resnick, N.M., Hayes, W.C.: Fall severity and
bone mineral density as risk factors for hip fracture in ambulatory elderly. JAMA 271(2),
128–133 (1994)
25. Hansen, U., Zioupos, P., Simpson, R., Currey, J.D., Hynd, D.: The effect of strain rate on the
mechanical properties of human cortical bone. J. Biomech. Eng. 130(1), 011011 (2008).
doi:10.1115/1.2838032
26. Hayes, W.C.: Biomechanics of cortical and trabecular bone: implications for assessment of
fracture risk. In: Mow, V.C., Hayes, W.C. (eds.) Basic Orthopaedic Biomechanics,
pp. 93–142. Raven Press, New York (1991)
27. Hayes, W.C., Myers, E.R., Morris, J.N., Gerhart, T.N., Yett, H.S., Lipsitz, L.A.: Impact near
the hip dominates fracture risk in elderly nursing home residents who fall. Calcif. Tissue Int.
52(3), 192–198 (1993)
28. Hayes, W.C., Myers, E.R., Robinovitch, S.N., Van Den Kroonenberg, A., Courtney, A.C.,
McMahon, T.A.: Etiology and prevention of age-related hip fractures. Bone 18(1 Suppl),
77S–86S (1996)
29. Hayes, W.C., Erickson, M.S., Power, E.D.: Forensic injury biomechanics. Annu. Rev.
Biomed. Eng. 9, 55–86 (2007). doi:10.1146/annurev.bioeng.9.060906.151946
30. Heller, M.O., Bergmann, G., Deuretzbacher, G., Durselen, L., Pohl, M., Claes, L., Haas, N.P.,
Duda, G.N.: Musculo-skeletal loading conditions at the hip during walking and stair
climbing. J. Biomech. 34(7), 883–893 (2001)
31. Ismail, A.A., Cooper, C., Felsenberg, D., Varlow, J., Kanis, J.A., Silman, A.J., O’Neill, T.W.:
Number and type of vertebral deformities: epidemiological characteristics and relation to
back pain and height loss. European Vertebral Osteoporosis Study Group. Osteoporos. Int.
9(3):206–213 (1999)
32. Kazakia, G.J., Burghardt, A.J., Link, T.M., Majumdar, S.: Variations in morphological and
biomechanical indices at the distal radius in subjects with identical BMD. J. Biomech.
(2010). doi:10.1016/j.jbiomech.2010.10.010
33. Keaveny, T.M.: Biomechanical computed tomography-noninvasive bone strength analysis
using clinical computed tomography scans. Ann. N. Y. Acad. Sci. 1192, 57–65 (2010).
doi:NYAS5348[pii]10.1111/j.1749-6632.2009.05348.x
34. Keaveny, T.M., Bouxsein, M.L.: Theoretical implications of the biomechanical fracture
threshold. J. Bone Miner. Res. 23(10), 1541–1547 (2008). doi:10.1359/jbmr.080406
35. Keyak, J.H., Rossi, S.A., Jones, K.A., Skinner, H.B.: Prediction of femoral fracture load using
automated finite element modeling. J. Biomech. 31(2), 125–133 (1998)
36. Klotzbuecher, C.M., Ross, P.D., Landsman, P.B., Abbott 3rd, T.A., Berger, M.: Patients with
prior fractures have an increased risk of future fractures: a summary of the literature and
statistical synthesis. J. Bone Miner. Res. 15(4), 721–739 (2000)
37. Lochmuller, E.M., Miller, P., Burklein, D., Wehr, U., Rambeck, W., Eckstein, F.: In situ
femoral dual-energy X-ray absorptiometry related to ash weight, bone size and density, and
its relationship with mechanical failure loads of the proximal femur. Osteoporos. Int. 11(4),
361–367 (2000)
38. Lochmuller, E.M., Groll, O., Kuhn, V., Eckstein, F.: Mechanical strength of the proximal
femur as predicted from geometric and densitometric bone properties at the lower limb versus
the distal radius. Bone 30(1), 207–216 (2002)
39. Lotz, J., Cheal, E., Hayes, W.: Stress distributions within the proximal femur during gait and
falls: Implications for osteoporotic fracture. Osteoporos. Int. 5, 252–261 (1995)
40. Mann, T., Oviatt, S.K., Wilson, D., Nelson, D., Orwoll, E.S.: Vertebral deformity in men.
J. Bone Miner. Res. 7(11), 1259–1265 (1992)
41. Marks, R., Allegrante, J.P., Ronald MacKenzie, C., Lane, J.M.: Hip fractures among the
elderly: causes, consequences and control. Ageing Res. Rev. 2(1), 57–93 (2003)
42. Matsumoto, T., Ohnishi, I., Bessho, M., Imai, K., Ohashi, S., Nakamura, K.: Prediction of
vertebral strength under loading conditions occurring in activities of daily living using a
computed tomography-based nonlinear finite element method. Spine 34(14), 1464–1469
(2009). doi:10.1097/BRS.0b013e3181a55636
43. Melton 3rd, L.J., Kan, S.H., Frye, M.A., Wahner, H.W., O’Fallon, W.M., Riggs, B.L.:
Epidemiology of vertebral fractures in women. Am. J. Epidemiol. 129(5), 1000–1011 (1989)
44. Melton 3rd, L.J., Riggs, B.L., Keaveny, T.M., Achenbach, S.J., Hoffmann, P.F., Camp, J.J.,
Rouleau, P.A., Bouxsein, M.L., Amin, S., Atkinson, E.J., Robb, R.A., Khosla, S.: Structural
determinants of vertebral fracture risk. J. Bone Miner. Res. 22(12), 1885–1892 (2007)
45. Melton 3rd, L.J., Riggs, B.L., van Lenthe, G.H., Achenbach, S.J., Muller, R., Bouxsein, M.L.,
Amin, S., Atkinson, E.J., Khosla, S.: Contribution of in vivo structural measurements and
load/strength ratios to the determination of forearm fracture risk in postmenopausal women.
J. Bone Miner. Res. 22(9), 1442–1448 (2007). doi:10.1359/jbmr.070514
46. Melton 3rd, L.J., Christen, D., Riggs, B.L., Achenbach, S.J., Muller, R., van Lenthe, G.H.,
Amin, S., Atkinson, E.J., Khosla, S.: Assessing forearm fracture risk in postmenopausal
women. Osteoporos. Int. 21(7), 1161–1169 (2010). doi:10.1007/s00198-009-1047-2
47. Melton 3rd, L.J., Riggs, B.L., Keaveny, T.M., Achenbach, S.J., Kopperdahl, D., Camp, J.J.,
Rouleau, P.A., Amin, S., Atkinson, E.J., Robb, R.A., Therneau, T.M., Khosla, S.: Relation of
vertebral deformities to bone density, structure, and strength. J. Bone Miner. Res. 25(9),
1922–1930 (2010). doi:10.1002/jbmr.150
48. Muller, M.E., Webber, C.E., Bouxsein, M.L.: Predicting the failure load of the distal radius.
Osteoporos. Int. 14(4), 345–352 (2003). doi:10.1007/s00198-003-1380-9
49. Myers, E.R., Wilson, S.E.: Biomechanics of osteoporosis and vertebral fracture. Spine
22(24 Suppl), 25S–31S (1997)
50. Myers, E.R., Robinovitch, S.N., Greenspan, S.L., Hayes, W.C.: Factor of risk is associated
with frequency of hip fracture in a case-control study. Trans. Orth. Res. Soc. 19, 526 (1994)
51. Myers, E., Wilson, S., Greenspan, S.: Vertebral fractures in the elderly occur with falling and
bending. J. Bone Miner. Res. 11, S355 (1996)
52. Nevitt. M.C., Cummings, S.R.: Type of fall and risk of hip and wrist fractures: the study of
osteoporotic fractures. The study of osteoporotic fractures research group. J. Am. Geriatr.
Soc. 41(11):1226—1234 (1993)
53. Nevitt, M.C., Ettinger, B., Black, D.M., Stone, K., Jamal, S.A., Ensrud, K., Segal, M., Genant, H.K.,
Cummings, S.R.: The association of radiographically detected vertebral fractures with back pain
and function: a prospective study. Ann. Intern. Med. 128(10), 793–800 (1998)
54. Nielson, C.M., Bouxsein, M.L., Freitas, S.S., Ensrud, K.E., Orwoll, E.S.: Trochanteric soft
tissue thickness and hip fracture in older men. J. Clin. Endocrinol. Metab. 94(2), 491–496
(2009). doi:10.1210/jc.2008-1640
55. Nordsletten, L., Ekeland, A.: Muscle contraction increases the structural capacity of the lower
leg: an in vivo study in the rat. J. Orthop. Res. 11(2), 299–304 (1993)
56. Orwoll, E.S., Marshall, L.M., Nielson, C.M., Cummings, S.R., Lapidus, J., Cauley, J.A.,
Ensrud, K., Lane, N., Hoffmann, P.R., Kopperdahl, D.L., Keaveny, T.M.: Finite element
analysis of the proximal femur and hip fracture risk in older men. J. Bone Miner. Res. 24(3),
475–483 (2009). doi:10.1359/jbmr.081201
57. OTA: Hip Fracture Outcomes in People Age 50 and Over-Background Paper. vol OTA-BP-
H- 120 U.S. Government Printing Office, Washington (1994)
58. Palvanen, M., Kannus, P., Parkkari, J., Pitkajarvi, T., Pasanen, M., Vuori, I., Jarvinen, M.:
The injury mechanisms of osteoporotic upper extremity fractures among older adults:
a controlled study of 287 consecutive patients and their 108 controls. Osteoporos. Int. 11(10),
822–831 (2000)
59. Pinilla, T.P., Boardman, K.C., Bouxsein, M.L., Myers, E.R., Hayes, W.C.: Impact direction
from a fall influences the failure load of the proximal femur as much as age-related bone loss.
60. Pistoia, W., van Rietbergen, B., Lochmuller, E.M., Lill, C.A., Eckstein, F., Ruegsegger, P.:
Estimation of distal radius failure load with micro-finite element analysis models based on
three-dimensional peripheral quantitative computed tomography images. Bone 30(6),
842–848 (2002). doi:S8756328202007366[pii]
61. Reilly, D.T., Burstein, A.H.: The elastic and ultimate properties of compact bone tissue.
J. Biomech. 8(6), 393–405 (1975)
62. Riggs, B.L., Melton 3rd, L.J., Robb, R.A., Camp, J.J., Atkinson, E.J., Peterson, J.M.,
Rouleau, P.A., McCollough, C.H., Bouxsein, M.L., Khosla, S.: Population-based study of age
and sex differences in bone volumetric density, size, geometry, and structure at different
skeletal sites. J. Bone Miner. Res. 19(12), 1945–1954 (2004). doi:10.1359/JBMR.040916
63. Riggs, B.L., Melton 3rd, L.J., Robb, R.A., Camp, J.J., Atkinson, E.J., Oberg, A.L., Rouleau,
P.A., McCollough, C.H., Khosla, S., Bouxsein, M.L.: Population-based analysis of the
relationship of whole bone strength indices and fall-related loads to age- and sex-specific
patterns of hip and wrist fractures. J. Bone Miner. Res. 21(2), 315–323 (2006). doi:10.1359/
JBMR.051022
64. Roberts, B.J., Thrall, E., Muller, J.A., Bouxsein, M.L.: Comparison of hip fracture risk
prediction by femoral aBMD to experimentally measured factor of risk. Bone 46(3), 742–746
(2010). doi:10.1016/j.bone.2009.10.020
65. Robinovitch, S.N., Hayes, W.C., McMahon, T.A.: Distribution of contact force during impact
to the hip. Ann. Biomed. Eng. 25(3), 499–508 (1997)
66. Robinovitch, S.N., Hayes, W.C., McMahon, T.A.: Predicting the impact response of a
nonlinear single-degree-of-freedom shock-absorbing system from the measured step
response. J. Biomech. Eng. 119(3), 221–227 (1997)
67. Schuit, S.C., van der Klift, M., Weel, A.E., de Laet, C.E., Burger, H., Seeman, E., Hofman, A.,
Uitterlinden, A.G., van Leeuwen, J.P., Pols, H.A.: Fracture incidence and association with bone
mineral density in elderly men and women: the Rotterdam Study. Bone 34(1), 195–202 (2004).
doi:S8756328203003776
68. Silva, M.J.: Biomechanics of osteoporotic fractures. Injury 38(Suppl 3), S69–S76 (2007).
doi:10.1016/j.injury.2007.08.014
69. Siris, E.S., Chen, Y.T., Abbott, T.A., Barrett-Connor, E., Miller, P.D., Wehren, L.E.,
Berger, M.L.: Bone mineral density thresholds for pharmacological intervention to prevent
fractures. Arch. Intern. Med. 164(10), 1108–1112 (2004). doi:10.1001/archinte.164.10.1108
70. Sornay-Rendu, E., Munoz, F., Garnero, P., Duboeuf, F., Delmas, P.D.: Identification of
osteopenic women at high risk of fracture: the OFELY study. J. Bone Miner. Res. 20(10),
1813–1819 (2005). doi:10.1359/JBMR.050609
71. Stone, K.L., Seeley, D.G., Lui, L.Y., Cauley, J.A., Ensrud, K., Browner, W.S., Nevitt, M.C.,
Cummings, S.R.: BMD at multiple sites and risk of fracture of multiple types: long-term
results from the Study of Osteoporotic Fractures. J. Bone Miner. Res. 18(11), 1947–1954
(2003). doi:10.1359/jbmr.2003.18.11.1947
72. Turner, C.H., Wang, T., Burr, D.B.: Shear strength and fatigue properties of human cortical
bone determined from pure shear tests. Calcif. Tissue Int. 69(6), 373–378 (2001)
73. van den Kroonenberg, A.J., Hayes, W.C., McMahon, T.A.: Dynamic models for sideways
falls from standing height. J. Biomech. Eng. 117(3), 309–318 (1995)
74. van den Kroonenberg, A.J., Hayes, W.C., McMahon, T.A.: Hip impact velocities and body
configurations for voluntary falls from standing height. J. Biomech. 29(6), 807–811 (1996).
doi:0021-9290(95)00134-4
75. Vilayphiou, N., Boutroy, S., Szulc, P., van Rietbergen, B., Munoz, F., Delmas, P.D.,
Chapurlat, R.: Finite element analysis performed on radius and tibia HR-pQCT images and
fragility fractures at all sites in men. J. Bone Miner. Res. 26(5), 965–973 (2011). doi:10.1002/
jbmr.297
76. Wainwright, S.A., Marshall, L.M., Ensrud, K.E., Cauley, J.A., Black, D.M., Hillier, T.A.,
Hochberg, M.C., Vogt, M.T., Orwoll, E.S.: Hip fracture in women without osteoporosis.
J. Clin. Endocrinol. Metab. 90(5), 2787–2793 (2005). doi:10.1210/jc.2004-1568
77. Yang, R.S., Liu, T.K., Hang, Y.S., Chieng, P.U., Tsai, K.S.: Factor of risk for hip fracture in
normal Chinese men and women in Taiwan. Calcif. Tissue Int. 65(6), 422–426 (1999).
doi:CTI-463
Bisphosphonates and PTH for Preventing
Fractures
David B. Burr and Matthew R. Allen
Abstract The risk of fracture is intimately linked to loss of bone mass. The two
most common pharmaceutical agents used to alter this loss are bisphosphonates
and recombinant human parathyroid hormone (rhPTH 1-34; teriparatide). These
two classes of drugs work through distinctly different mechanisms. Bisphos-
phonates bind to bone mineral and inhibit osteoclast activity. This leads to a
reduction in bone remodeling, which slows bone loss, and also leads to signif-
icant changes in the bone material properties such as mineralization, micro-
damage, and the organic matrix. The long-term effects of these altered material
properties are unclear. There are also noteworthy differences among the various
bisphosphonates used clinically such as the speed of onset and magnitude of
remodeling suppression. PTH is an anabolic agent which stimulates both
remodeling and modeling—resulting in the formation of new bone which over
time leads to an increase in bone mass. PTH also alters the material properties
although these changes are distinctly different from the bisphosphonates. Recent
studies have begun to investigate combining bisphosphonates and PTH, either
sequentially or concomitantly, with most data showing that bisphosphonates
D. B. Burr (&) M. R. Allen

Department of Anatomy and Cell Biology, Indiana University School of Medicine,
Indianapolis, IN, USA
e-mail: dburr@iupui.edu
M. R. Allen
e-mail: matallen@iupui.edu
D. B. Burr
Department of Biomedical Engineering,
Indiana University-Purdue University at Indianapolis (IUPUI),
Indianapolis, IN 46077, USA

DOI: 10.1007/8415_2011_81
Published Online: 21 April 2011
152 D. B. Burr and M. R. Allen
blunt the full effect of PTH. Although bisphosphonates and PTH each have their
own specific profile, mechanisms of action, and effects on bone mass, archi-
tecture and tissue properties, both have been shown to effectively reduce ver-
tebral and non-vertebral fractures and improve the health of postmenopausal
women and older men.
1 Introduction
At the most basic level, there are only two ways to alter bone mass and thus reduce
the risk of fractures caused by osteoporosis (Fig. 1). One way—the anti-catabolic
route—is to prevent the loss of bone that accompanies aging and the reduction in
sex hormones. The other—the anabolic route—is to increase bone mass through
net bone formation.
Bisphosphonates (BPs) are the most common anti-catabolic agents whether
taken daily, weekly, monthly, or yearly—orally or as an intravenous infusion–BPs
significantly decrease fracture incidence at vertebral and non-vertebral sites, as
documented in several clinical trials of postmenopausal osteoporotic women
which have been summarized elsewhere [1]. Similar efficacy exists for reducing
fractures in other conditions associated with increased fracture risk such as aged
men, glucocorticoid-induced osteoporosis, Paget’s disease, osteogenesis imperfect
and cancer patients. The most widely studied condition where BPs have shown
fracture risk reduction is post-menopausal women where vertebral and non-ver-
tebral fracture risk is almost universally reduced over 1–3 years of treatment—the
duration of most clinical trials. In general BPs reduce fracture risk by *60% in
post-menopausal women at the vertebra, hip, and non-vertebral sites. The success
of the BPs in reducing fracture risk is generally attributed to their suppression of
bone remodeling, which maintains (or minimally increases) bone mass as well as
increasing its mineralization. Together these changes typically result in bone
mineral density (BMD) increasing by 3–12% [1, 2].
The only current anabolic agent approved for the treatment of osteoporosis is
recombinant human parathyroid hormone, PTH(1-34), or teriparatide. Taken once
daily as a subcutaneous injection, teriparatide is generally used for women with
extremely low BMD who need a therapy that can significantly increase bone mass
rather than simply reducing loss. Teriparatide significantly reduces fracture risk at
both vertebral and non-vertebral sites in post-menopausal women [3], and is used
to treat other conditions such as hypophosphatasia [4]. The success of teriparatide
in reducing fracture risk is attributed to its ability to increase the amount of bone
by stimulating direct apposition of new bone to trabecular surfaces [5–7] and by
allowing overfilling of resorption spaces at each bone remodeling site [8]. This
typically results in BMD increases of 10–15% [3, 9] over 18–24 months of
treatment.
Bisphosphonates and PTH for Preventing Fractures 153
Fig. 1 Agents that act

primarily by reducing bone
loss are classified as anti-
catabolic while those that act
by altering bone valance in
favor of formation, or cause
direct apposition to bone
surfaces, are anabolic
2 The Mechanism of Action of Bisphosphonates
Bisphosphonates work by suppressing activation of osteoclasts and inducing their

apoptosis by interfering with the mevalonate pathway [1, 10]. Upon liberation
from the bone surface in the low pH environment of resorption lacunae, BPs are
taken up by the osteoclast through the ruffled boarder via fluid-phase endocy-
tosis. Once in the cell, BPs inhibit both FPPS (farnesyl diphosphonate synthase)
and GGPPS (geranylgeranyl diphosphate synthase), blocking prenylation of
small GTPases such as Ras, Rho, Rac and Rap as well as cell survival signaling
pathways. Inhibiting the interaction of these GTPases with the cell membrane
down-regulates signaling through the Akt and ERK 1/2 signaling pathways,
causing the release of cytochrome c, and eventually the activation of caspases.
BPs also acts by perturbing the cell cycle of osteoclasts, inhibiting cell growth
by inducing cell cycle arrest in the S-phase. The cellular effects of BPs are
largely confined to osteoclasts, as these are the cells most often subjected to the
highest concentrations of drug in vivo. Recently, the clinical observation of an
acute phase reaction following intravenous BP treatment led to the discovery that
other cells (gamma delta T cells in the case of the acute phase reaction, but also
macrophages) can be adversely affected in vivo by BPs if they see sufficient
concentrations [11–14]. This emphasizes the point that BPs do not specifically
target osteoclasts, and any cell that sees sufficiently high concentrations is likely
to have disrupted function.
By suppressing the initiation of osteoclast activity, BPs reduce the number of
active bone multicellular units (BMUs) and the erosion pits that do form are not as
large. In untreated individuals there normally exists a net negative bone balance
within each BMU meaning a small amount of bone loss occurs at the BMU-level
with each remodeling cycle. In post-menopausal women, there is an increase in
bone remodeling sites, and also a larger net bone loss within each BMU—together
these are responsible for the accelerated bone loss during the post-menopausal
years. Although some differences exist in the degree to which activation frequency
(the variable assessed histologically to determine the degree of remodeling) is
suppressed, BP-induced suppression usually exceeds 70% in cancellous bone,
assessed in iliac crest biopsies. This suppression of remodeling significantly
Fig. 2 Higher bone strength with bisphosphonate treatment is explained by the drugs’ effect on
bone density. Following 1, 2, or 3 years of treatment with daily oral alendronate in dogs (total
n = 84), the ultimate load of the lumbar vertebrae is significantly increased. This higher bone
strength is almost entirely explained by the higher areal bone mineral density (aBMD) as the
relationship between ultimate load and aBMD is not different from vehicle-treated animals (total
n = 36)
reduces the loss of bone that occurs through remodeling and allows sites that are
actively remodeling to refill. This latter effect is responsible for the initial increase
in bone mass that is usually observed with BP treatment. In addition to reducing
the number of BMUs BPs also have recently been shown to decrease the size of
those few BMUs that are initiated [15]. As BPs have minimal effects on osteoblast
activity, this reduction in BMU size means that even those sites that do undergo
remodeling in the presence of BP treatment lose less bone than would normally
occur.
Reductions of remodeling and preservation (or small increases) in bone mass
translate into improved whole bone mechanical properties—the ultimate goal of
any bone treatment. Virtually all of the effects of the BPs on bone strength, and
presumably fracture resistance, can be attributed to increased BMD achieved
through maintenance of the amount of bone and increases in its mineralization
(Fig. 2) [16]. Following 1–3 years of treatment with clinical doses of BPs, the
relationship between vertebral ultimate load, a measure of strength, and areal
BMD was nearly identical in animals treated with BP or with saline vehicle. That
is, the increased compressive strength in BP-treated animals was entirely
accounted for by increased BMD, and at a given BMD, BP and control treated
animals had similar bone strength. Clinical data corroborate this relationship, at
least to a point, showing that those BPs that increase BMD the most, reduce
fracture to the greatest degree (Table 1). This is important as some individuals
treated with BPs do not see large increases in BMD—therefore these patients may
have reduced fracture protection.
Table 1 Increased vertebral BMD and reduced fracture incidence after 3 years treatment with
different bisphosphonates 6
Compound Increased BMD Reduction in fracture indicies Suppression of turnover5
Alendronate1 6.2% 50% 92%
Risedronate2 5.4% 41% 40%
Ibandronate3 6.5% 50% 50%
Zoledronate4 6.7% 70% 63%
1
See references [134, 135]
2
See reference [136]; this varies depending on the dose
3
See reference [137]
4
See reference [138]
5
See text for references
6
For comparison of vertebral fracture rates among bisphosphonate treatments, see [139]
3 The Effects of BPs on Remodeling Suppression
Binding affinity of the drug can have a significant effect on the accumulation of
the BP within the bone matrix, the speed with which its effect is initiated, the
reversibility of effect once the drug is withdrawn [17], as well as diffusion of
the drug into the bone [12, 14] and potential recycling of BP released consequent
to bone remodeling [18]. Differences in the mineral binding affinities of BPs used
clinically are, in rank order from highest mineral affinity, with their affinity con-
stants (KL 9 106) [19, 20]:
zoledronate ð3:9Þ [ alendronate ð2:9Þ [ ibandronate ð2:3Þ [ risedronate ð2:0Þ
Consistent with the affinity constants, approximately 50% more risedronate is

excreted than alendronate after 24 h [21, 22], resulting in significantly greater
skeletal retention of alendronate compared to risedronate after 28 days.
BPs also demonstrate different potencies in their effect on osteoclast FPPS.
Differences in FPPS inhibition of bisphosphonates used clinically are, in rank
order from the highest to lowest [23, 24]:
zoledronate [ risedronate [ ibandronate [ alendronate
It is these two properties—binding affinity and potency on osteoclasts–along
with the dose and route of administration–that determine the physiological effects
on bone remodeling which are similar in some respects yet differ with regard to the
magnitude of suppression, speed of onset of effect, and skeletal retention.
3.1 Magnitude and Site Specificity of Remodeling Suppression
Bisphosphonates accumulate in the skeleton in a dose-dependent manner [25–27].

Treatment of dogs with a wide range of risedronate doses (including the post-
menopasual osteoporosis dose) show a dose–response in vertebral [28] and iliac
Fig. 3 Effects of
bisphosphonates on iliac crest
activation frequency.
Activation frequency, a
histological variable used to
evaluate the rate of bone
remodeling, has been
assessed in separate clinical
trials. These data show
significantly lower values in
BP-treated patients compared
to placebo controls in each of
the four trials. See text for
individual study references
crest [29] activation frequency. No such dose–response was shown in a similar

analysis using doses of alendronate in dogs [28]. Ovariectomized non-human
primates showed a dose–response in remodeling suppression after 18 months of
treatment with ibandronate [30]. Human studies hint at a dose–response in
remodeling suppression of the iliac crest in post-menopausal women treated with
alendronate yet this was only observed at doses below those used clinically [31].
Alendronate significantly suppresses remodeling more than risedronate [32, 33]
and zoledronate suppresses more than risedronate [34]. There are no head-to-head
assessments of bone remodeling using histology, yet in separate studies the percent
suppression of remodeling relative to placebo controls tends to be relatively
similar over 3 years with daily risedronate [35], daily alendronate [36], intermit-
tent oral ibandronate [37], and intravenous yearly zoledronate [38] (Fig. 3).
The effect of BPs on bone turnover is highly location specific and when
assessed histologically can differ by an order of magnitude across skeletal sites.
There is also a time-dependent effect on remodeling suppression. Remodeling was
reduced only 15% in the mandible after 6 months of treatment with clinical doses
of alendronate [39], but treatment for three years at these same doses reduced
alveolar remodeling by 67% compared to vehicle-treated controls [40]. The same
time-dependent suppression was shown in the rib where clinical doses had no
effect after one year but reduced remodeling by more than 85% in the following
three years with osteoporosis treatment doses [41, 42]. Clearly, the effect of BP
treatment on remodeling suppression is dose, time, and location dependent and
also differs among the specific BPs.
3.2 On and Off Effect of Remodeling Suppression
Animal studies show that normal trabecular bone remodeling tends to be

re-established sooner in animals previously treated with risedronate compared to
those previously treated with alendronate [17]. Animals treated with clinically
relevant doses of either risedronate or alendronate for 8 weeks and assessed
16 weeks after discontinuation of treatment, show a return to control values in
risedronate treated animals but not in alendronate treated animals. These differ-
ences in recovery of remodeling following treatment withdrawal were ascribed to
differences in binding affinity.
Differences in binding affinity, and in potency, may also be reflected in how
quickly initial administration of bisphosphonates begin to have their effects.
Recent data shows that the incidence of non-vertebral fractures and the incidence
of hip fractures in the first year on therapy is significantly lower in risedronate-
treated patients than in alendronate-treated patients [43]. This concept was recently
addressed in an animal model in which, as early as three weeks after the initiation
of treatment with clinically relevant doses, vertebral bone remodeling was sup-
pressed to a significantly greater degree in risedronate-treated animals than
alendronate-treated animals. All BPs have a semi-unique remodeling suppression
fingerprint and this likely plays a role in their clinical efficacy.
3.3 Long-Term Suppression of Remodeling
The high affinity of BPs for bone mineral, and their long-term retention in bone,
are of some concern because continued accumulation of BPs, or continued sup-
pression of remodeling over prolonged treatment periods could eventually increase
the risk of fracture, even in the face of increased bone mass. The seven-year
alendronate clinical trial data show an increase in the annualized incidence of new
vertebral fractures rates in years 6 and 7 compared to baseline placebo group
fracture rates. In the first three years of the clinical trial [44], the placebo group
sustained an annualized vertebral fracture incidence of 2.1%, compared to 0.9% in
the alendronate treated group. During years 6 and 7 of the two year extension
study, the alendronate treated group sustained an annualized vertebral fracture
incidence of 3.3%, more than 50% greater than the incidence of vertebral fractures
in the placebo group at the beginning of the study [45]. This is not an exact
comparison because of differences in methods of assessing vertebral fractures, the
absence of a placebo control in years 6 and 7 (due to ethical considerations once
efficacy is shown), and the fact that the mean age of the women in the extension
study was undoubtedly older than at the initiation of the trial, although the mean
age is not reported in the extension study. This raises concerns that long-term
treatment with BPs could ultimately be detrimental to the health of the patient.
Recently, several groups have reported an apparent increase in the incidence of
atypical femoral fractures, especially in the population of osteopenic women who
are being treated for osteoporosis with alendronate for an average of 4–8 years
[46–49]. Although epidemiologic data on atypical femoral fractures is not
extensive (and the terminology used to describe femoral fractures is often con-
fusing and inconsistent), low energy subtrochanteric fractures are not infrequent in
postmenopausal women. Estimates place the prevalence without bisphosphonate

treatment at about 7/100,000 person-years in women 55–74 years old, and 34–74/
100,000 person-years in women older than 75 years [50]. However, atpypical
femoral fractures—defined as low energy fractures that result in a transverse or
oblique fracture orientation, with a medial spike and a lateral periosteal stress
reaction–may only account for about one-third of all subtrochanteric fractures,
many of which are spiral and not transverse [51, 52].
The investigators found that only 7% of patients who presented with atypical
femoral fractures had been exposed to alendronate, the same percentage as those
with a typical hip fracture. As alendronate is known to reduce the risk for hip
fracture, this would suggest that alendronate use does not contribute to atypical
femoral fractures. The risk associated with ‘‘subtrochanteric/diaphyseal’’ femoral
fractures in patients on alendronate was 1.46, compared to 1.45 for hip fracture,
after adjustment for co-morbidity and co-medications. The overall conclusion is
that atypical femoral fractures are probably another brand of osteoporotic fracture,
and that fracture risk is not increased by the use of alendronate. A more recent
study [53] that re-analyzed subtrochanteric/diaphyseal fractures from three large
clinical trials of BPs suggest calculated relative hazard ratios from 1.03 for those
treated with alendronate, to 1.5 for those women treated with zoledronate. They
conclude that treatment for three years with alendronate would prevent 10-fold
more hip fractures than it would cause, and that treatment even for 10 years does
not pose a significant risk for atypical femoral fractures.
The concept of BPs increasing fracture risk is certainly counterintuitive given
these agents are prescribed for, and known to be effective in, reducing fractures.
However, BPs exert well-known effects on bone tissue that tend to make the tissue
more brittle (see Sect. 4). It is possible that in the short term, increases in BMD
offset negative changes in the tissue level, yet over the long term (or in some
patients) the detrimental effects on the tissue reach a point that even with a higher
BMD the patient fractures.
4 The Effects of BPs on Bone Matrix Properties
BPs are effective in reducing the risk of fracture in postmenopausal women at least
over 10 years [54], but numerous studies demonstrate that they negatively affect
bone tissue quality. This means that the benefit of BPs for reducing fracture is due
primarily to maintenance/small increases in bone mass and volume. BP treatment
is associated with increases in bone mineralization, microdamage, and alterations
in the cross-linking of collagen (Fig. 4). Any of these changes, either alone or in
combination, could potentially compromise the mechanical properties of the tis-
sue. Pre-clinical studies have shown consistently that the mechanical properties of
the tissue, specifically material toughness (the normalized energy to fracture) are
reduced with BP-treatment. Following 1–3 years of treatment at doses at or above
those used in postmenopausal women, bone toughness is 20–30% lower compared
Fig. 4 Overview of
determinants of structural
biomechanical properties and
how bisphosphonates affect
the key material properties of
bone
to control animals [28, 55–57]. This decline in toughness was initially thought to
be related to the well-documented accumulation of microdamage that was observed
in lumbar vertebrae and other bones in dogs treated with BPs [28, 56, 57], although
changes to both mineralization and collagen cross-linking have also been shown to
occur. More recent data show that toughness in BP-treated animals continues to
decline with long-term treatment without a change in microdamage accumulation
or a further increase in mineralization [55]. This suggests that neither microdamage
nor mineralization is completely responsible for the progressive deterioration in the
bone’s material properties leaving progressive changes to collagen, or the inter-
action among all these properties, as the cause of this progressive toughness
decline.
4.1 Mineralization
Both pre-clinical and clinical studies show that by reducing the turnover of bone
and thereby increasing mean tissue age, BP treatments lead to a significantly
higher average tissue mineralization [58, 59] and lower heterogeneity of miner-
alization across the bone matrix [60]. These changes probably occur predomi-
nately within the first 2–3 years of treatment, and then change little with continued
treatment [61, 62]. Increased mineralization with BP treatment occurs for two
reasons. Under normal conditions, bone remodeling preferentially renews the more
highly mineralized bone matrix, a process that takes about a year [63] or longer
[64]. Thus by suppressing remodeling, BPs allow more highly mineralized regions
to persist for a longer time. Moreover, suppression of remodeling allows more of
the newly formed bone to become fully mineralized without replacement.
An unresolved question is whether BPs alter either the rate of mineralization, or
the eventual degree of mineralization, of a specific BMU. One or both of these
changes would be expected to have a significant effect on the biomechanical
properties at the tissue level. Early data using FTIR [59] imply that BPs may
acutely slow the initial rate of primary mineralization within the first month, but
that subsequent secondary mineralization occurs normally and allows the final
level of mineralization to be equivalent to that of untreated bone. However, more
recently, Fuchs et al. showed that administration of risedronate or alendronate had
no effect on the rate of either primary or secondary mineralization [65]. While both
drugs increase the overall mineralization of the tissue by suppressing remodeling
and allowing more sites to achieve full mineralization, they don’t alter the rate at
which full mineralization is achieved. There also was no significant effect after one
year of treatment on the final level of BMU mineralization, suggesting that hyper
mineralization at the BMU does not occur.
Increased mineralization will increase both the strength and the stiffness of
bone—an important design goal for reducing the risk of fracture—but increased
material stiffness is inevitably associated with reduced energy absorption
(toughness) [66]. Post-yield stress and strain are also compromised by increasing
levels of mineralization [67]. In this regard, the 8–10% increases in mean degree
of mineralization reported following 2–3 years of alendronate treatment could be
cause for concern. On the other hand, more recent studies in animals [68] show no
relationship between small (*2%) increases in overall tissue mineralization
(percent ash weight) that occur with 1–3 years of bisphosphonate treatment, and
toughness.
4.2 Collagen and the Organic Matrix
Bone collagen contains both enzymatic and non-enzymatic collagen crosslinks that
stablize the matrix and have significant impact on the bone’s mechanical prop-
erties. The organic matrix constitutes the principal toughening mechanism in bone,
and therefore plays a substantial role in determining properties of energy
absorption/toughness [69]. Changes in the organic matrix may have some effect on
pre-yield tissue strength and stiffness [70, 71], although these properties are pre-
dominantly determined by the mineral fraction. Cross-links formed through non-
enzymatic processes are associated with tissue that is more brittle [72], and has
reduced post-yield deformation [73, 74], work to fracture [75, 76], and toughness
[77].
Following one-year of treatment with a wide-range of BP doses, the ratio of
pyridinoline to deoxypyridinoline (PYD/DPD, an index of increasing cross-link
maturity) in the trabecular bone of lumbar vertebrae was significantly increased
compared to vehicle-treated animals. The level of pentosidine, an advanced
glycation end-product (AGE) was significantly increased in vertebral trabecular
bone and cortical bone of the tibia from bisphosphonate treatment animals com-
pared to controls [75, 77]. In a separate experiment, levels of pentosidine were
found to be increased in the rib of dogs following 3 years of treatment with
incadronate [78]. Limited data exist assessing collagen crosslinks in humans
treated with BPs. Using FTIR (Fourier Transformed Infrared Spectroscopy) BPs
had no effect on collagen maturity in iliac crest biopsies [79].
4.3 Microdamage
Bisphosphonate treatment is associated with changes in both the initiation and

repair of microdamage. This has been demonstrated repeatedly in animal models
using various different oral bisphosphonates at doses ranging from to 69 the
dose used for treatment of post-menopausal osteoporosis, and with treatment
durations lasting from 1 to 3 years. Significantly higher levels of microdamage are
consistently noted in the trabecular bone of the lumbar vertebrae and cortical bone
of the rib with bisphosphonate treatment [80]. Although increased levels of mi-
crodamage have also been noted in the ilium, thoracic spinous process, and
femoral neck of dogs treated with bisphosphonates, these sites appear less prone to
significant microdamage accumulation (\2-fold relative to untreated) [57, 80].
This site-specificity may be important in evaluating damage accumulation in bone
from human patients; as such evaluations can only occur from iliac crest biopsies
which may underestimate the amount of damage accumulating in the spine or ribs.
Recent data from iliac crest biopsies of treatment naïve women and women treated
for 5 years with alendronate show increased microdamage accumulation with
bisphosphonate treatment [81]. Both low femoral neck BMD and increasing age
were associated with greater microdamage formation, suggesting that older
patients with especially low BMD might be more at risk for damage accumulation.
A separate study in which iliac crest biopsies of women treated with alendronate
were compared to cadaveric bone showed no significant difference in micro-
damage levels, although in this study the cadavers used as ‘‘untreated controls’
were almost 10 years older than the treated patients [82]. The well-known age-
related increase in microdamage [83–85], therefore, may make this an unsuitable
control population. Moreover, there was no independent verification that the
control population had not used bisphosphonates while alive.
The increased brittleness caused by changes to bone’s organic matrix and
mineralization allow for greater initiation of microdamage [80]. In the majority of
studies that have documented increased microdamage with BP-treatment, a con-
comitant decrease in bone toughness has also been quantified [28, 55–57, 86].
However, recent studies assessing microdamage and biomechanical properties in
dog vertebra suggest microdamage accumulation may not be the predominant
reason for reduced toughness [55]. In a one year study using various doses of
alendronate or risedronate, there was minimal congruence between changes in
microdamage accumulation and material-level toughness in vertebrae from several
groups of bisphosphonate-treated dogs [28]. Although these data do not eliminate
the possibility of a direct cause/effect connection, they suggest factors other than
microdamage contribute significantly to the material-level biomechanical changes
associated with bisphosphonate treatment. This conclusion is supported by data
Fig. 5 Discordant changes in microdamage accumulation and vertebral toughness following one
or 3 years of bisphosphonate treatment in beagle dogs. Following one year of treatment with
various doses of daily oral risedronate (R) or alendronate (A), vertebral microdamage was
significantly higher compared to vehicle (VEH). In these same animals, toughness was
consistently lower in all bisphosphonate-treated groups. In animals treated with A or VEH for
3 years, microdamage was significantly higher in all groups compared to 1 year VEH animals yet
toughness was only reduced in the A-treated groups. These data clearly illustrate that increases in
microdamage are not universally associated with reductions in toughness, but that bisphosphonate
treatment is likely the key factor associated with reductions in toughness
showing that non-BP treated animals that have an age-related 2-fold increase in
microdamage accumulation had no change in bone toughness [55] (Fig. 5). Thus
the current theory is that microdamage accumulation with BPs is more likely the
consequence of the increased brittleness and reduced toughness, and not the cause
of it. AGEs naturally accumulate in bone as it ages, but under normal rates of bone
turnover, they are prevented from accumulating to high levels. However, when
bone turnover is suppressed, they can accumulate and make it more likely for
cracks to initiate.
5 The Mechanism of Action of Recombinant

Human Parathyroid Hormone (1-34)
Recombinant human parathyroid hormone [rhPTH (1-34), or teriparatide] is the

only anabolic agent for bone approved for use in humans. Another analogue of
PTH (PTH-1-84) has also been developed for treatment of osteoporosis. This drug
is currently approved in Europe, but not in the U.S. because its effect on bone was
essentially similar to that of teriparatide, and studies did not demonstrate superi-
ority [87–89].
Parathyroid hormone causes an early and direct apposition of lamellar bone

within the first several weeks following the initiation of treatment [90, 91]. At the
same time, it may increase the trabecular and endocortical bone surfaces activation
frequency for remodeling, causing a transient increase in porosity at sites com-
posed of cortical bone. This can lead to a temporary reduction in BMD at those
sites, which has been observed most notably at the hip [92]. Increased remodeling
also occurs in cancellous bone, but new sites of bone resorption on trabeculae
eventually overfill, with bone formation extending beyond the margins of the
erosion cavity [6–8], leading to a net gain in bone even at those remodeling sites.
These observations led to the concept of an anabolic window in which the early
increase in bone formation and volume, and presumably strength, was not offset by
the initiation of bone resorption as a consequence of increased activation fre-
quency [93, 94]. The anabolic window occurs even with prior exposure to
alendronate or raloxifene [95].
Initially, there was concern that the early increase in bone remodeling would
weaken an already osteoporotic skeleton, and that even though this was temporary,
it could lead to an early increase in fracture risk within the first 6 months of
treatment. This, however, has not been observed clinically. No investigator has
reported an increased fracture incidence within the first 1 to 3 months of treatment
even though many women on teriparatide therapy have extremely low bone mass.
And longer-term studies clearly demonstrate anti-fracture efficacy at non-vertebral
sites in postmenopausal women with osteoporosis [96]. Moreover, although there
is an initial decline in BMD at the hip within the first 6 months of treatment, BMD
increases at the hip during the 2 year of treatment [97]. This suggests that the
increase in porosity is either a transient phenomenon, or that apposition of bone to
trabecular and endocortical surfaces more than compensates for the intracortical
loss after the first year.
Following 18–24 months of treatment with teriparatide, BMD in the spine is
increased by 10–15% [3, 9]. The reduction in risk of fracture with teriparatide
treatment is comparable to that found with bisphosphonate administration. Nine-
teen months of treatment with teriparatide reduced the relative risk of spine
fracture by 65% (absolute risk: 9.3%), and nonvertebral fragility fractures by 53%
(absolute risk: 2.9%), but no significant reduction in hip fractures [3], similar to the
findings with PTH (1-84) [88].
6 The Effect of Teriparatide on Geometry and Architecture
6.1 Cortical Bone
Teriparatide increases cortical thickness, but whether apposition of bone to the

periosteal surface contributes to this is controversial. rhPTH (1-34) stimulates both
periosteal and endocortical bone formation in animals [98–101], but this was not
found in a nonhuman primate model [102]. A periosteal effect has also been
difficult to verify in humans because of the resolution of imaging systems and the
variability inherent in histomorphometric analyses of iliac crest biopsies. Iliac
crest biopsy studies have consistently shown increased cortical thickness with PTH
[5, 8, 103] but have not documented increased bone formation rate on either
periosteal or endocortical surfaces with PTH treatment [5].
Using pQCT, an increase in periosteal circumference of the distal radius was
documented in women treated for postmenopausal osteoporosis for 18 months
with either 20 lg or 40 lg daily treatments of teriparatide [104]. They also found
an increase in endocortical circumference at the higher dose. These changes did
not alter cortical thickness, but the bending and torsional rigidities of the radius
were greater because of the larger periosteal diameter. However, the analysis was
cross-sectional, so it is difficult to determine whether this was a real increase, or
the result of sampling. In a separate study, longitudinal measures of the intertro-
chanteric and neck regions of the femur were made over a two year treatment
program [105]. This study was unable to detect significant periosteal expansion
compared to placebo-treated patients, but did detect a reduction in the marrow
cavity diameter, suggesting that bone formation along the endocortical surface
may account for the majority of increased cortical thickness in humans.
6.2 Trabecular Bone
Trabecular and endocortical surfaces are the primary locations of teriparatide’s

activity. The consequence of this is to significantly increase trabecular bone vol-
ume and connectivity, through a sequential increase in trabecular thickness that
transforms trabecular rods into plates [8, 103] followed by architectural modifi-
cations that increase trabecular number. The initial direct apposition of bone to
trabecular surfaces leads to thicker trabeculae, and because this occurs quickly,
provides an immediate protective mechanical benefit. Over time, these thick tra-
beculae are remodeled and often bisected by the cutting fronts of new remodeling
units, through a process now known as trabecular ‘‘tunneling’’ [106]. This process
normalizes trabecular thickness, and creates additional surfaces for direct bone
formation, allowing for more rapid increase in bone volume through surface
apposition. Increased trabecular thickness and/or number each enhance mechani-
cal properties independent of increases in bone volume [107, 108]. However,
changes in trabecular number can affect strength twice as much as changes in
trabecular thickness [109] (Fig. 6). Thus the improved trabecular connectivity that
occurs as the result of tunneling is an effective long-term strategy to increase the
density and strength of bone much more than could be accomplished by trabecular
bone apposition alone. The conversion from a rod-like to a plate-like morphology
also creates a structure that is better able to withstand load bearing in variable
directions, and resists buckling that can occur in very thin rods with high aspect
ratios. Such architectural changes could increase stiffness of trabeculae even with
Fig. 6 Hypothetical effect of

changing trabecular bone in
response to teriparatide
treatment. Based on
mathematical modeling of
Silva et al. [109] which
determined the mechanical
consequences of reducing
BV/TV by either decreasing
trabecular thickness or
number, the increase in
mechanical properties
achieved by teriparatide are
optimized by increasing the
trabecular number as opposed
to altering thickness
smaller improvements in bone volume fraction simply by altering the curvature of

the trabecular surfaces [110], a beneficial effect that would be most pronounced
when the bone volume fraction is low.
7 Effects of Teriparatide on Bone Matrix Properties
Teriparatide has a significant and rapid effect on increasing bone mass, but by
increasing remodeling rate, it also renews aging tissue. Treatment with teriparatide
results in the replacement of older, more highly mineralized bone with younger,
less mineralized tissue, resulting in greater tissue heterogeneity [111–113]. This
also results in a lower mineralization density, with less highly crystalline
hydroxyapatite. Moreover, it results in a higher ratio of divalent to trivalent cross-
links in collagen, the direct result of a higher turnover rate. In cynomolgus
monkeys treated with two different doses of rhPTH (1-34), the changes in tissue
properties were fully reversible at the lower dose, but were sustained for at least
two remodeling periods after withdrawal in the animals given the higher dose
[114]. This is consistent with the effects being closely tied to remodeling rate. The
stimulation of remodeling by PTH not only alters tissue properties, but also
removes excess microdamage [115], and would create additional interfaces that
arrest the growth of new microcracks. The mechanical effect of each of these
changes individually is not clear, but in combination the result is a toughening
effect, making the tissue more compliant and allowing the bone to absorb more
energy prior to fracture [102, 116].
8 Combining Bisphosphonates and Teriparatide
Teriparatide will promote a significant improvement in bone mass and stimulation

of bone formation, even in those patients who have been pre-treated with other
drugs for years [117–119]. However, most of the evidence available suggests that
administration of bisphosphonates prior to teriparatide treatment [95], or concur-
rent administration of both [97, 120], will impair or at least delay the full anabolic
response to PTH. In two landmarks studies [97, 120], investigators found that bone
mineral density increased more in those given teriparatide alone, than in those
given both alendronate and teriparatide, although the combination treatment still
performed better than alendronate alone. These effects appear to last for at least
6–12 months [95, 121]. This is not the case for all anti-catabolic agents; PTH’s
effects are not blunted by prior administration with raloxifene, a selective estrogen
receptor modulator (SERM) [95, 122] or by hormone replacement therapy [123].
Moreover, the blunting effect varies even among different BPs. Administration of
teriparatide for one year had a greater effect on bone formation, measured by
serum biomarkers, and BMD of the spine in patients treated previously for 2 years
with risedronate than in those treated previously with alendronate [118], with a
concomitant increase in predicted stiffness and failure load [124]. The difference
between effects with prior risedronate treatment compared to prior alendronate
treatment were apparent as early as 3 months following the initiation of teri-
paratide therapy [118]. These differences in PTH responsiveness among the BPs
are likely related to differences in their binding affinity.
The results from studies using animal models provide a mixed picture. There
was no effect of prior treatment with alendronate on trabecular bone formation in
ovariectomized rats [125], whereas others [126] do detect a suppression of skeletal
responsiveness to PTH following alendronate pretreatment. Again, the experience
with bisphosphonates that have different binding affinities varies in this regard.
A recent study in aged rats also showed that an 8 week period of risedronate
treatment blunted the stimulatory effect of subsequent PTH on bone remodeling
[127], but bone formation by PTH was not suppressed 10 weeks after withdrawal
of risedronate administration.
The blunting effect of bisphosphonate pre-treatment on PTH stimulation of
bone formation may diminish over treatment time. In the EUROFORS study, the
initial significant delay in the increase of BMD [128], as a surrogate for bone
formation, in the group treated with alendronate, was completely reversed
following 2 years of PTH therapy, resulting in a significant increase in BMD
[117, 129]. Using iliac crest biopsies, Stepan et al. similarly found no difference in
bone formation rates following 19 months of teriparatide treatment between those
who were previously treated with alendronate for 5 years, and those who were not
previously treated [119]. The microdamage accumulation caused by remodeling
suppression was reversed by subsequent administration of teriparatide, and those
patients with the lowest BMD had the greatest removal of microdamage [115].
Although this study was not able to conclude that a change in bone turnover rate
was associated with this reduction in microdamage, such a conclusion seems

eminently plausible.
Notably, even though the co-administration or prior administration of a bis-
phosphonate will blunt the subsequent effect of PTH on BMD, the response of co-
treatment is still better than switching entirely to PTH once the bisphosphonate
therapy has begun. Adding teriparatide to ongoing therapy with either alendronate
or raloxifene improves BMD more than stopping the anti-resorptive therapy and
switching to PTH [122]. Co-administration of an anti-resorptive (either raloxifene
or alendronate) with PTH (1-84) still provides greater increases in hip BMD than
did PTH (1-84) alone [120, 130]. Whether this is true for other skeletal sites is not
clear; these results may be both a function of the regulation of the PTH-stimulated
effects on bone remodeling specifically at the hip, and the short period of PTH
administration (6 months).
Concurrent administration of PTH with a BP does not appear to generate either
additive or synergistic effects. To some extent, any advantage appears dominated
by the role of BPs. Thus, increases in spine BMD are not greater with co-
administration of PTH + alendronate than either drug alone, and BMD at the hip is
greater in the groups treated with alendronate, whether in combination with PTH
or not, than with administration of PTH alone [120, 131]. However, when PTH
treatment is followed by administration of alendronate, BMD continues to increase
in the spine [131]. A meta-analysis of combination treatments showed that PTH
treatment either with or without co-administration with a BP reduced the risk of
vertebral fracture by 64%, and of non-vertebral fracture by 38%, compared to 48
and 49% respectively for alendronate alone. This would suggest that PTH alone or
in combination has a greater positive benefit on the spine, whereas administration
of a BP alone has a better profile for the hip. However, these results are prob-
lematic in that they pool studies in which inclusion criteria, dose and duration
vary.
Current treatment regulations only allow administration of teriparatide for up to
2 years, because of a perceived risk of osteosarcoma that was identified in pre-
clinical trials in rats which received lifetime administration of PTH [132]. How-
ever, cessation of treatment with PTH will, within 18–30 months [96, 133], result
in a reduction in BMD and an increase in vertebral fracture risk. Subsequent
treatment with a BP after full exposure to teriparatide will stabilize or increase
BMD and reduce fracture risk [96].
9 Conclusion
Bisphosphonates are the most widely used anti-catabolic agents to prevent frac-
tures in various forms of osteoporosis, and to prevent metastasis to bone in certain
kinds of cancer. Teriparatide (rhPTH 1-34) currently is the only anabolic agent
available to treat osteoporosis. Both classes of therapy have their own specific
profile, mechanisms of action, and effects on bone mass, architecture and tissue
properties, but both have been shown to effectively reduce vertebral and non-
vertebral fractures. Newer agents that have different mechanisms of action have
either been approved recently (e.g., denosumab, an anti-catabolic) or may be
available soon (e.g., anti-sclerostin antibody, an anabolic). However, both the bis-
phosphonates and parathyroid hormone have been very effective in reducing risk of
fracture and improving the health of both postmenopausal women and older men.
References
1. Russell, R.G., Watts, N.B., Ebetino, F.H., Rogers, M.J.: Mechanisms of action of
bisphosphonates: similarities and differences and their potential influence on clinical
efficacy. Osteoporos. Int. 19, 733–759 (2008)
2. Delmas, P.D., Seeman, E.: Changes in bone mineral density explain little of the reduction in
vertebral or nonvertebral fracture risk with anti-resorptive therapy. Bone 34, 599–604 (2004)
3. Neer, R.M., Arnaud, C.D., Zanchetta, J.R., Prince, R., Gaich, G.A., Reginster, J.Y.,
Hodsman, A.B., Eriksen, E.F., Ish-Shalom, S., Genant, H.K., Wang, O., Mitlak, B.H.: Effect
of parathyroid hormone(1-34) on fractures and bone mineral density in postmenopausal
women with osteoporosis. N. Engl. J. Med. 344, 1434–1441 (2001)
4. Whyte, M.P., Mumm, S., Deal, C.: Adult hypophosphatasia treated with teriparatide. J. Clin.
5. Arlot, M., Meunier, P.J., Boivin, G., Haddock, L., Tamayo, J., Correa-Rotter, R., Jasqui, S.,
Donley, D.W., Dalsky, G.P., Martin, J.S., Eriksen, E.F.: Differential effects of teriparatide
and alendronate on bone remodeling in postmenopausal women assessed by
histomorphometric parameters. J. Bone Miner. Res. 20, 1244–1253 (2005)
6. Lindsay, R., Cosman, F., Zhou, H., Bostrom, M.P., Shen, V.W., Cruz, J.D., Nieves, J.W.,
Dempster, D.W.: A novel tetracycline labeling schedule for longitudinal evaluation of the
short-term effects of anabolic therapy with a single iliac crest bone biopsy: early actions of
teriparatide. J. Bone Miner. Res. 21, 366–373 (2006)
7. Lindsay, R., Zhou, H., Cosman, F., Nieves, J., Dempster, D.W., Hodsman, A.B.: Effects of a
one-month treatment with PTH(1-34) on bone formation on cancellous, endocortical, and
periosteal surfaces of the human ilium. J. Bone Miner. Res. 22, 495–502 (2007)
8. Dempster, D.W., Cosman, F., Kurland, E.S., Zhou, H., Nieves, J., Woelfert, L., Shane, E.,
Plavetic, K., Muller, R., Bilezikian, J., Lindsay, R.: Effects of daily treatment with
parathyroid hormone on bone microarchitecture and turnover in patients with osteoporosis:
a paired biopsy study. J. Bone Miner. Res. 16, 1846–1853 (2001)
9. Stroup, J.S., Rivers, S.M., Abu-Baker, A.M., Kane, M.P.: Two-year changes in bone
mineral density and T scores in patients treated at a pharmacist-run teriparatide clinic.
Pharmacotherapy 27, 779–788 (2007)
10. Rodan, G.A., Fleisch, H.A.: Bisphosphonates: mechanisms of action. J. Clin. Invest. 97,
2692–2696 (1996)
11. Galluzzo, S., Santini, D., Vincenzi, B., Caccamo, N., Meraviglia, F., Salerno, A., Dieli, F.,
Tonini, G.: Immunomodulating role of bisphosphonates on human gamma delta T cells: an
intriguing and promising aspect of their antitumour activity. Expert. Opin. Ther. Targets 11,
941–954 (2007)
12. Roelofs, A.J., Coxon, F.P., Ebetino, F.H., Lundy, M.W., Henneman, Z.J., Nancollas, G.H.,
Sun, S., Blazewska, K.M., Lynn, F.B.J., Kashemirov, B.A., Khalid, A.B., McKenna, C.E.,
Rogers, M.J.: Fluorescent risedronate analogs reveal bisphosphonate uptake by bone
marrow monocytes and localization around osteocytes in vivo. J. Bone Miner. Res. 12, 12
(2009)
13. Coxon, F., Thompson, K., Ebetino, H., Rogers, M.: Resorbing osteoclasts increase the
availability of mineral-bound bisphosphoantes to non-resorbing cells. Bone 38, S45 (2006)
14. Coxon, F.P., Thompson, K., Roelofs, A.J., Ebetino, F.H., Rogers, M.J.: Visualizing mineral
binding and uptake of bisphosphonate by osteoclasts and non-resorbing cells. Bone 42, 848–
860 (2008)
15. Allen, M.R., Erickson, A.M., Wang, X., Burr, D.B., Martin, R.B., Hazelwood, S.J.:
Morphological assessment of basic multicellular unit resorption parameters in dogs shows
additional mechanisms of bisphosphonate effects on bone. Calcif. Tissue Int. 86, 67–71
(2010)
16. Allen, M.R., Burr, D.B.: Changes in vertebral strength-density and energy absorption-
density relationships following bisphosphonate treatment in beagle dogs. Osteoporos. Int.
19, 95–99 (2008)
17. Fuchs, R.K., Phipps, R.J., Burr, D.B.: Recovery of trabecular and cortical bone turnover
following discontinuation of risedronate and alendronate therapy in ovariectomized rats.
J. Bone Miner. Res. 23, 1689–1697 (2008)
18. Sato, M., Grasser, W., Endo, N., Akins, R., Simmons, H., Thompson, D.D., Golub, E.,
Rodan, G.A.: Bisphosphonate action - alendronate localization in rat bone and effects on
osteoclast ultrastructure. J. Clin. Invest. 88, 2095–2105 (1991)
19. Nancollas, G.H., Tang, R., Phipps, R.J., Henneman, Z., Gulde, S., Wu, W., Mangood, A.,
Russell, R.G., Ebetino, F.H.: Novel insights into actions of bisphosphonates on bone:
differences in interactions with hydroxyapatite. Bone 38, 617–627 (2005)
20. Leu, C.T., Luegmayr, E., Freedman, L.P., Rodan, G.A., Reszka, A.A.: Relative binding
affinities of bisphosphonates for human bone and relationship to antiresorptive efficacy.
Bone 38, 628–636 (2006)
21. Khan, S.A., Kanis, J.A., Vasikaran, S., Kline, W.F., Matuszewski, B.K., McCloskey, E.V.,
Beneton, M.N., Gertz, B.J., Sciberras, D.G., Holland, S.D., Orgee, J., Coombes, G.M.,
Rogers, S.R., Porras, A.G.: Elimination and biochemical responses to intravenous
alendronate in postmenopausal osteoporosis. J. Bone Miner. Res. 12, 1700–1707 (1997)
22. Mitchell, D.Y., Barr, W.H., Eusebio, R.A., Stevens, K.A., Duke, F.P., Russell, D.A.,
Nesbitt, J.D., Powell, J.H., Thompson, G.A.: Risedronate pharmacokinetics and intra- and
inter-subject variability upon single-dose intravenous and oral administration. Pharm. Res.
18, 166–170 (2001)
23. Dunford, J.E., Thompson, K., Coxon, F.P., Luckman, S.P., Hahn, F.M., Poulter, C.D.,
Ebetino, F.H., Rogers, M.J.: Structure-activity relationships for inhibition of farnesyl
diphosphate synthase in vitro and inhibition of bone resorption in vivo by nitrogen-
containing bisphosphonates. J. Pharmacol. Exp. Ther. 296, 235–242 (2001)
24. Dunford, J.E., Kwaasi, A.A., Rogers, M.J., Barnett, B.L., Ebetino, F.H., Russell, R.G.,
Oppermann, U., Kavanagh, K.L., Watts, N.B.: Structure-activity relationships among the
nitrogen containing bisphosphonates in clinical use and other analogues: time-dependent
inhibition of human farnesyl pyrophosphate synthase. J. Med. Chem. 51, 2187–2195 (2008)
25. Lin, J.H.: Bisphosphonates: a review of their pharmacokinetic properties. Bone 18, 75–85
(1996)
26. Lin, J.H., Chen, I.W., deLuna, F.A.: Nonlinear kinetics of alendronate. Plasma protein
binding and bone uptake. Drug Metab. Dispos. 22, 400–405 (1994)
27. Lin, J.H., Chen, I.W., Duggan, D.E.: Effects of dose, sex, and age on the disposition of
alendronate, a potent antiosteolytic bisphosphonate, in rats. Drug Metab. Dispos. 20, 473–
478 (1992)
28. Allen, M.R., Iwata, K., Phipps, R., Burr, D.B.: Alterations in canine vertebral bone turnover,
microdamage accumulation, and biomechanical properties following 1-year treatment with
clinical treatment doses of risedronate or alendronate. Bone 39, 872–879 (2006)
29. Boyce, R.W., Paddock, C.L., Gleason, J.R., Sletsema, W.K., Eriksen, E.F.: The effects of
risedronate on canine cancellous bone remodeling: three-dimensional kinetic reconstruction
of the remodeling site. J. Bone Miner. Res. 10, 211–221 (1995)
30. Smith, S.Y., Recker, R.R., Hannan, M., Muller, R., Bauss, F.: Intermittent intravenous
administration of the bisphosphonate ibandronate prevents bone loss and maintains bone
strength and quality in ovariectomized cynomolgus monkeys. Bone 32, 45–55 (2003)
31. Bone, H.G., Downs Jr., R.W., Tucci, J.R., Harris, S.T., Weinstein, R.S., Licata, A.A.,
McClung, M.R., Kimmel, D.B., Gertz, B.J., Hale, E., Polvino, W.J.: Dose-response
relationships for alendronate treatment in osteoporotic elderly women. Alendronate elderly
osteoporosis study centers. J. Clin. Endocrinol. Metab. 82, 265–274 (1997)
32. Bonnick, S., Saag, K.G., Kiel, D.P., McClung, M., Hochberg, M., Burnett, S.M., Sebba, A.,
Kagan, R., Chen, E., Thompson, D.E., de Papp, A.E.: Comparison of weekly treatment of
postmenopausal osteoporosis with alendronate versus risedronate over two years. J. Clin.
33. Rosen, C.J., Hochberg, M.C., Bonnick, S.L., McClung, M., Miller, P., Broy, S., Kagan, R.,
Chen, E., Petruschke, R.A., Thompson, D.E., de Papp, A.E.: Treatment with once-weekly
alendronate 70 mg compared with once-weekly risedronate 35 mg in women with
postmenopausal osteoporosis: a randomized double-blind study. J. Bone Miner. Res. 20,
141–151 (2005)
34. Reid, I.R., Miller, P., Lyles, K., Fraser, W., Brown, J.P., Saidi, Y., Mesenbrink, P., Su, G.,
Pak, J., Zelenakas, K., Luchi, M., Richardson, P., Hosking, D.: Comparison of a single
infusion of zoledronic acid with risedronate for Paget’s disease. N. Engl. J. Med. 353, 898–
908 (2005)
35. Borah, B., Ritman, E.L., Dufresne, T.E., Jorgensen, S.M., Liu, S., Sacha, J., Phipps, R.J.,
Turner, R.T.: The effect of risedronate on bone mineralization as measured by micro-
computed tomography with synchrotron radiation: correlation to histomorphometric indices
of turnover. Bone 37, 1–9 (2005)
36. Chavassieux, P.M., Arlot, M.E., Reda, C., Wei, L., Yates, A.J., Meunier, P.J.:
Histomorphometric assessment of the long-term effects of alendronate on bone quality
and remodeling in patients with osteoporosis. J. Clin. Invest. 100, 1475–1480 (1997)
37. Recker, R.R., Weinstein, R.S., Chesnut 3rd, C.H., Schimmer, R.C., Mahoney, P., Hughes, C.,
Bonvoisin, B., Meunier, P.J.: Histomorphometric evaluation of daily and intermittent oral
ibandronate in women with postmenopausal osteoporosis: results from the BONE study.
Osteoporos. Int. 15, 231–237 (2004)
38. Recker, R.R., Delmas, P.D., Halse, J., Reid, I.R., Boonen, S., Garcia-Hernandez, P.A.,
Supronik, J., Lewiecki, E.M., Ochoa, L., Miller, P., Hu, H., Mesenbrink, P., Hartl, F.,
Gasser, J., Eriksen, E.F.: Effects of intravenous zoledronic acid once yearly on bone
remodeling and bone structure. J. Bone Miner. Res. 23, 6–16 (2008)
39. Allen, M.R., Kubek, D.J., Burr, D.B.: Cancer treatment dosing regimens of zoledronic acid
result in near-complete suppression of mandible intracortical bone remodeling in beagle
dogs. J. Bone Miner. Res. 25, 98–105 (2010)
40. Allen, M.R., Burr, D.B.: Mandible matrix necrosis in beagle dogs after 3 years of daily oral
bisphosphonate treatment. J. Oral. Maxillofac. Surg. 66, 987–994 (2008)
41. Allen, M.R., Follet, H., Khurana, M., Sato, M., Burr, D.B.: Antiremodeling agents influence
osteoblast activity differently in modeling and remodeling sites of canine rib. Calcif. Tissue
Int. 79, 255–261 (2006)
42. Allen, M.R., Reinwald, S., Burr, D.B.: Alendronate reduces bone toughness of ribs without
significantly increasing microdamage accumulation in dogs following 3 years of daily
treatment. Calcif. Tissue Int. 82, 354–360 (2008)
43. Silverman, S.L., Watts, N.B., Delmas, P.D., Lange, J.L., Lindsay, R.: Effectiveness of
bisphosphonates on nonvertebral and hip fractures in the first year of therapy: the
risedronate and alendronate (REAL) cohort study. Osteoporos. Int. 18, 25–34 (2007)
44. Liberman, U.A., Weiss, S.R., Broll, J., Minne, H.W., Quan, H., Bell, N.H., Rodriguez-
Portales, J., Downs Jr., R.W., Dequeker, J., Favus, M.: Effect of oral alendronate on bone
mineral density and the incidence of fractures in postmenopausal osteoporosis. The
alendronate phase III osteoporosis treatment study group. N. Engl. J. Med. 333, 1437–1443
(1995)
45. Tonino, R.P., Meunier, P.J., Emkey, R., Rodriguez-Portales, J.A., Menkes, C.J., Wasnich, R.D.,
Bone, H.G., Santora, A.C., Wu, M., Desai, R., Ross, P.D.: Skeletal benefits of alendronate:
7-year treatment of postmenopausal osteoporotic women. Phase III osteoporosis treatment study
group. J. Clin. Endocrinol. Metab. 85, 3109–3115 (2000)
46. Goh, S.K., Yang, K.Y., Koh, J.S., Wong, M.K., Chua, S.Y., Chua, D.T., Howe, T.S.:
Subtrochanteric insufficiency fractures in patients on alendronate therapy: a caution. J. Bone
Joint Surg. Br. 89, 349–353 (2007)
47. Lenart, B.A., Lorich, D.G., Lane, J.M.: Atypical fractures of the femoral diaphysis in
postmenopausal women taking alendronate. N. Engl. J. Med. 358, 1304–1306 (2008)
48. Capeci, C.M., Tejwani, N.C.: Bilateral low-energy simultaneous or sequential femoral
fractures in patients on long-term alendronate therapy. J. Bone Joint Surg. Am. 91, 2556–
2561 (2009)
49. Neviaser, A.S., Lane, J.M., Lenart, B.A., Edobor-Osula, F., Lorich, D.G.: Low-energy
femoral shaft fractures associated with alendronate use. J. Orthop. Trauma 22, 346–350
(2008)
50. Arneson, T.J., Melton 3rd, L.J., Lewallen, D.G., O’Fallon, W.M.: Epidemiology of
diaphyseal and distal femoral fractures in Rochester, Minnesota, 1965–1984. Clin. Orthop.
Relat. Res. 234, 188–194 (1988)
51. Salminen, S., Pihlajamaki, H., Avikainen, V., Kyro, A., Bostman, O.: Specific features
associated with femoral shaft fractures caused by low-energy trauma. J. Trauma 43, 117–
122 (1997)
52. Salminen, S.T., Pihlajamaki, H.K., Avikainen, V.J., Bostman, O.M.: Population based
epidemiologic and morphologic study of femoral shaft fractures. Clin. Orthop. Relat. Res.
372, 241–249 (2000)
53. Black, D.M., Kelly, M.P., Genant, H.K., Palermo, L., Eastell, R., Bucci-Rechtweg, C.,
Cauley, J., Leung, P.C., Boonen, S., Santora, A., de Papp, A., Bauer, D.C.: Bisphosphonates
and fractures of the subtrochanteric or diaphyseal femur. N. Engl J. Med. 362, 1761–1771
(2010)
54. Bone, H.G., Hosking, D., Devogelaer, J.-P., Tucci, J.R., Emkey, R.D., Tonino, R.P.,
Rodriguez-Portales, J.A., Downs, R.W., Gupta, J., Santora, A.C., Liberman, U.A.: The
alendronate phase III osteoporosis treatment study group. Ten years’ experience with
alendronate for osteoporosis in postmenopausal women. N. Engl. J. Med. 350, 1189–1199
(2004)
55. Allen, M.R., Burr, D.B.: Three years of alendronate treatment results in similar levels of
vertebral microdamage as after one year of treatment. J. Bone Miner. Res. 22, 1759–1765
(2007)
56. Mashiba, T., Hirano, T., Turner, C.H., Forwood, M.R., Johnston, C.C., Burr, D.B.:
Suppressed bone turnover by bisphosphonates increases microdamage accumulation and
reduces some biomechanical properties in dog rib. J. Bone Miner. Res. 15, 613–620 (2000)
57. Mashiba, T., Turner, C.H., Hirano, T., Forwood, M.R., Johnston, C.C., Burr, D.B.: Effects
of suppressed bone turnover by bisphosphonates on microdamage accumulation and
biomechanical properties in clinically relevant skeletal sites in beagles. Bone 28, 524–531
(2001)
58. Boivin, G.Y., Chavassieux, P.M., Santora, A.C., Yates, J., Meunier, P.J.: Alendronate
increases bone strength by increasing the mean degree of mineralization of bone tissue in
osteoporotic women. Bone 27, 687–694 (2000)
59. Burr, D.B., Miller, L., Grynpas, M., Li, J.L., Boyde, A., Mashiba, T., Hirano, T., Johnston,
C.C.: Tissue mineralization is increased following 1-year treatment with high doses of
bisphosphonates in dogs. Bone 33, 960–969 (2003)
60. Roschger, P., Rinnerthaler, S., Yates, J., Rodan, G.A., Fratzl, P., Klaushofer, K.:
Alendronate increases degree and uniformity of mineralization in cancellous bone and
decreases the porosity in cortical bone of osteoporotic women. Bone 29, 185–191 (2001)
61. Borah, B., Dufresne, T.E., Ritman, E.L., Jorgensen, S.M., Liu, S., Chmielewski, P.A.,
Phipps, R.J., Zhou, X., Sibonga, J.D., Turner, R.T.: Long-term risedronate treatment
normalizes mineralization and continues to preserve trabecular architecture: sequential

triple biopsy studies with micro-computed tomography. Bone 39, 345–352 (2006)
62. Zoehrer, R., Roschger, P., Paschalis, E.P., Hofstaetter, J.G., Durchschlag, E., Fratzl, P.,
Phipps, R., Klaushofer, K.: Effects of 3- and 5-year treatment with risedronate on bone
mineralization density distribution in triple biopsies of the iliac crest in postmenopausal
women. J. Bone Miner. Res. 21, 1106–1112 (2006)
63. Fuchs, R.K., Allen, M.R., Ruppel, M.E., Diab, T., Phipps, R.J., Miller, L.M., Burr, D.B.: In
situ examination of the time-course for secondary mineralization of Haversian bone using
synchrotron Fourier transform infrared micro spectroscopy. Matrix. Biol. 27, 34–41 (2008)
64. Bala, Y., Farlay, D., Delmas, P.D., Meunier, P.J., Boivin, G.: Time sequence of secondary
mineralization and microhardness in cortical and cancellous bone from ewes. Bone 46,
1204–1212 (2010)
65. Fuchs, R.K., Allen, M.R., Ruppel, M.E., Miller, L.M., Phipps, R., Burr, D.B.: Do
bisphosphonates affect the rate of secondary mineralization of trabecular bone? J. Bone
Miner. Res. (2009) Suppl
66. Currey, J.: Incompatible mechanical properties in compact bone. J. Theor. Biol. 231, 569–
580 (2004)
67. Les, C.M., Stover, S.M., Keyak, J.H., Taylor, K.T., Kaneps, A.J.: Stiff and strong
compressive properties are associated with brittle post-yield behavior in equine compact
bone material. J. Orthop. Res. 20, 607–614 (2002)
68. Allen, M., Burr, D.: Skeletal microdamage: less about biomechanics and more about
remodeling. Clin. Rev. Bone Miner. Metab. 6, 24–30 (2008)
69. Wang, X., Shen, X., Li, X., Agrawal, C.M.: Age-related changes in the collagen network
and toughness of bone. Bone 31, 1–7 (2002)
70. Zioupos, P., Currey, J.D., Hamer, A.J.: The role of collagen in the declining mechanical
properties of aging human cortical bone. J. Biomed. Mater. Res. 45, 108–116 (1999)
71. Garnero, P., Borel, O., Gineyts, E., Duboeuf, F., Solberg, H., Bouxsein, M.L., Christiansen, C.,
Delmas, P.D.: Extracellular post-translational modifications of collagen are major
determinants of biomechanical properties of fetal bovine cortical bone. Bone 38, 300–309
(2006)
72. Vashishth, D.: Advanced glycation end-products and bone fractures. BoneKEy 6, 268–278
(2009)
73. Vashishth, D., Gibson, G.J., Khoury, J.I., Schaffler, M.B., Kimura, J., Fyhrie, D.P.: Influence
of nonenzymatic glycation on biomechanical properties of cortical bone. Bone 28, 195–201
(2001)
74. Tang, S.Y., Zeenath, U., Vashishth, D.: Effects of non-enzymatic glycation on cancellous
bone fragility. Bone 40, 1144–1151 (2007)
75. Tang, S.Y., Allen, M.R., Phipps, R., Burr, D.B., Vashishth, D.: Changes in non-enzymatic
glycation and its association with altered mechanical properties following 1-year treatment
with risedronate or alendronate. Osteoporos. Int. 20, 887–894 (2009)
76. Viguet-Carrin, S., Roux, J.P., Arlot, M.E., Merabet, Z., Leeming, D.J., Byrjalsen, I.,
Delmas, P.D., Bouxsein, M.L.: Contribution of the advanced glycation end product
pentosidine and of maturation of type I collagen to compressive biomechanical properties of
human lumbar vertebrae. Bone 39, 1073–1079 (2006)
77. Allen, M.R., Gineyts, E., Leeming, D.J., Burr, D.B., Delmas, P.D.: Bisphosphonates alter
trabecular bone collagen cross-linking and isomerization in beagle dog vertebra.
Osteoporos. Int. 19, 329–337 (2008)
78. Saito, M., Mori, S., Mashiba, T., Komatsubara, S., Marumo, K.: Collagen maturity,
glycation induced-pentosidine, and mineralization are increased following 3-year treatment
with incadronate in dogs. Osteoporos. Int. 19, 1343–1354 (2008)
79. Durchschlag, E., Paschalis, E.P., Zoehrer, R., Roschger, P., Fratzl, P., Recker, R., Phipps, R.,
Klaushofer, K.: Bone material properties in trabecular bone from human iliac crest
biopsies after 3- and 5-year treatment with risedronate. J. Bone Miner. Res. 21, 1581–1590
(2006)
80. Allen, M.R., Burr, D.B. Mineralization, microdamage, and matrix: how bisphosphonates
influence material properties of bone. BoneKEy 2007;4:49–60; http://www.bonekey-ibms.
org/cgi/content/abstract/ibmske;4/2/49
81. Stepan, J.J., Burr, D.B., Pavo, I., Sipos, A., Michalska, D., Li, J., Fahrleitner-Pammer, A.,
Petto, H., Westmore, M., Michalsky, D., Sato, M., Dobnig, H.: Low bone mineral density is
associated with bone microdamage accumulation in postmenopausal women with
osteoporosis. Bone 41, 378–385 (2007)
82. Chapurlat, R.D., Arlot, M., Burt-Pichat, B., Chavassieux, P., Roux, J.P., Portero-Muzy, N.,
Delmas, P.D.: Microcrack frequency and bone remodeling in postmenopausal osteoporotic
women on long-term bisphosphonates: a bone biopsy study. J. Bone Miner. Res. 22, 1502–
1509 (2007)
83. Diab, T., Condon, K.W., Burr, D.B., Vashishth, D.: Age-related change in the damage
morphology of human cortical bone and its role in bone fragility. Bone 38, 427–431 (2006)
84. Norman, T.L., Wang, Z.: Microdamage of human cortical bone: incidence and morphology
in long bones. Bone 20, 375–379 (1997)
85. Schaffler, M.B., Choi, K., Milgrom, C.: Aging and matrix microdamage accumulation in
human compact bone. Bone 17, 521–525 (1995)
86. Komatsubara, S., Mori, S., Mashiba, T., Ito, M., Li, J., Kaji, Y., Akiyama, T., Miyamoto, K.,
Cao, Y., Kawanishi, J., Norimatsu, H.: Long-term treatment of incadronate disodium
accumulates microdamage but improves the trabecular bone microarchitecture in dog
vertebra. J. Bone Miner. Res. 18, 512–520 (2003)
87. Sellmeyer, D.E., Black, D.M., Palermo, L., Greenspan, S., Ensrud, K., Bilezikian, J., Rosen, C.J.:
Hetereogeneity in skeletal response to full-length parathyroid hormone in the treatment of
osteoporosis. Osteoporos. Int. 18, 973–979 (2007)
88. Greenspan, S.L., Bone, H.G., Ettinger, M.P., Hanley, D.A., Lindsay, R., Zanchetta, J.R.,
Blosch, C.M., Mathisen, A.L., Morris, S.A., Marriott, T.B.: Effect of recombinant human
parathyroid hormone(1-84) on vertebral fracture and bone mineral density in
postmenopausal women with osteoporosis: a randomized trial. Ann. Intern. Med. 146,
326–339 (2007)
89. Recker, R.R., Bare, S.P., Smith, S.Y., Varela, A., Miller, M.A., Morris, S.A., Fox, J.:
Cancellous and cortical bone architecture and turnover at the iliac crest of postmenopausal
osteoporotic women treated with parathyroid hormone1-84. Bone 44, 113–119 (2009)
90. Hodsman, A.B., Fraher, L.J., Ostbye, T., Adachi, J.D., Steer, B.M.: An evaluation of several
biochemical markers for bone formation and resorption in a protocol utilizing cyclical
parathyroid hormone and calcitonin therapy for osteoporosis. J. Clin. Invest. 91, 1138–1148
(1993)
91. Hodsman, A.B., Kisiel, M., Adachi, J.D., Fraher, L.J., Watson, P.H.: Histomorphometric
evidence for increased bone turnover without change in cortical thickness or porosity after
2 years of cyclical hPTH(1-34) therapy in women with severe osteoporosis. Bone 27, 311–
318 (2000)
92. Lindsay, R., Nieves, J., Formica, C., Henneman, E., Woelfert, L., Shen, V., Dempster, D.,
Cosman, F.: Randomised controlled study of effect of parathyroid hormone on vertebral-
bone mass and fracture incidence among postmenopausal women on oestrogen with
osteoporosis. Lancet 350, 550–555 (1997)
93. Burr, D.B.: Does early PTH treatment compromise bone strength? The balance between
remodeling, porosity, bone mineral, and bone size. Curr. Osteoporos. Rep. 3, 19–24 (2005)
94. Rubin, M.R., Bilezikian, J.P.: New anabolic therapies in osteoporosis. Curr. Opin.
Rheumatol. 14, 433–440 (2002)
95. Ettinger, B., San Martin, J., Crans, G., Pavo, I.: Differential effects of teriparatide on BMD
after treatment with raloxifene or alendronate. J. Bone Miner. Res. 19, 745–751 (2004)
96. Prince, R., Sipos, A., Hossain, A., Syversen, U., Ish-Shalom, S., Marcinowska, E., Halse, J.,
Lindsay, R., Dalsky, G.P., Mitlak, B.H.: Sustained nonvertebral fragility fracture risk
reduction after discontinuation of teriparatide treatment. J. Bone Miner. Res. 20, 1507–1513
(2005)
97. Finkelstein, J.S., Hayes, A., Hunzelman, J.L., Wyland, J.J., Lee, H., Neer, R.M.: The effects
of parathyroid hormone, alendronate, or both in men with osteoporosis. N. Engl. J. Med.
349, 1216–1226 (2003)
98. Burr, D.B., Hirano, T., Turner, C.H., Hotchkiss, C., Brommage, R., Hock, J.M.: Intermittently
administered human parathyroid hormone (1-34) treatment increases intracortical bone
turnover and porosity without reducing bone strength in the humerus of ovariectomized
cynomolgus monkeys. J. Bone Miner. Res. 16, 157–165 (2001)
99. Hirano, T., Burr, D.B., Cain, R.L., Hock, J.M.: Changes in geometry and cortical porosity in
adult, ovary-intact rabbits after 5 months treatment with LY333334 (hPTH1-34). Calcif.
Tissue Int. 66, 456–460 (2000)
100. Hirano, T., Burr, D.B., Turner, C.H., Sato, M., Cain, R.L., Hock, J.M.: Anabolic effects of
human biosynthetic parathyroid hormone fragment(1-34), LY333334, on remodeling and
mechanical properties of cortical bone in rabbits. J. Bone Miner. Res. 14, 536–545 (1999)
101. Mashiba, T., Burr, D.B., Turner, C.H., Sato, M., Cain, R.L., Hock, J.M.: Effects of human
parathyroid hormone(1-34), LY333334, on bone mass, remodeling, and mechanical
properties of cortical bone during the first remodeling cycle in rabbits. Bone 28, 538–547
(2001)
102. Sato, M., Westmore, M., Ma, Y.L., Schmidt, A., Zeng, Q.Q., Glass, E.V., Vahle, J.,
Brommage, R., Jerome, C.P., Turner, C.H.: Teriparatide [PTH(1-34)] strengthens the
proximal femur of ovariectomized nonhuman primates despite increasing porosity. J. Bone
Miner. Res. 19, 623–629 (2004)
103. Jiang, Y., Zhao, J.J., Mitlak, B.H., Wang, O., Genant, H.K., Eriksen, E.F.: Recombinant
human parathyroid hormone(1-34) [teriparatide] improves both cortical and cancellous bone
structure. J. Bone Miner. Res. 18, 1932–1941 (2003)
104. Zanchetta, J.R., Bogado, C.E., Ferretti, J.L., Wang, O., Wilson, M.G., Sato, M., Gaich, G.A.,
Dalsky, G.P., Myers, S.L.: Effects of teriparatide [recombinant human parathyroid hormone
(1-34)] on cortical bone in postmenopausal women with osteoporosis. J. Bone Miner. Res. 18,
539–543 (2003)
105. Uusi-Rasi, K., Semanick, L.M., Zanchetta, J.R., Bogado, C.E., Eriksen, E.F., Sato, M.,
Beck, T.J.: Effects of teriparatide [rhPTH(1-34)] treatment on structural geometry of the
proximal femur in elderly osteoporotic women. Bone 36, 948–958 (2005)
106. Jerome, C.P., Burr, D.B., Van Bibber, T., Hock, J.M., Brommage, R.: Treatment with
human parathyroid hormone(1-34) for 18 months increases cancellous bone volume and
improves trabecular architecture in ovariectomized cynomolgus monkeys (Macaca
fascicularis). Bone 28, 150–159 (2001)
107. Guo, X.E., Kim, C.H.: Mechanical consequence of trabecular bone loss and its treatment: a
three-dimensional model simulation. Bone 30, 404–411 (2002)
108. Pistoia, W., van Rietbergen, B., Ruegsegger, P.: Mechanical consequences of different
scenarios for simulated bone atrophy and recovery in the distal radius. Bone 33, 937–945
(2003)
109. Silva, M.J., Gibson, L.J.: Modeling the mechanical behavior of vertebral trabecular bone:
effects of age-related changes in microstructure. Bone 21, 191–199 (1997)
110. Miller, Z., Fuchs, M.B.: Effect of trabecular curvature on the stiffness of trabecular bone.
J. Biomech. 38, 1855–1864 (2005)
111. Paschalis, E.P., Glass, E.V., Donley, D.W., Eriksen, E.F.: Bone mineral and collagen quality
in iliac crest biopsies of patients given teriparatide: new results from the fracture prevention
trial. J. Clin. Endocrinol. Metab. 90, 4644–4649 (2005)
112. Misof, B.M., Roschger, P., Cosman, F., Kurland, E.S., Tesch, W., Messmer, P., Dempster, D.W.,
Nieves, J., Shane, E., Fratzl, P., Klaushofer, K., Bilezikian, J., Lindsay, R.: Effects of intermittent
parathyroid hormone administration on bone mineralization density in iliac crest biopsies from
patients with osteoporosis: a paired study before and after treatment. J. Clin. Endocrinol. Metab.
88, 1150–1156 (2003)
113. Roschger, P., Paschalis, E.P., Fratzl, P., Klaushofer, K.: Bone mineralization density
distribution in health and disease. Bone 42, 456–466 (2008)
114. Paschalis, E.P., Burr, D.B., Mendelsohn, R., Hock, J.M., Boskey, A.L.: Bone mineral and
collagen quality in humeri of ovariectomized cynomolgus monkeys given rhPTH(1-34) for
18 months. J. Bone Miner. Res 18, 769–775 (2003)
115. Dobnig, H., Stepan, J.J., Burr, D.B., Li, J., Michalska, D., Sipos, A., Petto, H., Fahrleitner-
Pammer, A., Pavo, I.: Teriparatide reduces bone microdamage accumulation in
postmenopausal women previously treated with alendronate. J. Bone Miner. Res. 24,
1998–2006 (2009)
116. Ejersted, C., Andreassen, T.T., Hauge, E.M., Melsen, F., Oxlund, H.: Parathyroid
hormone(1-34) increases vertebral bone mass, compressive strength, and quality in old
rats. Bone 17, 507–511 (1995)
117. Boonen, S., Marin, F., Obermayer-Pietsch, B., Simoes, M.E., Barker, C., Glass, E.V., Hadji, P.,
Lyritis, G., Oertel, H., Nickelsen, T., McCloskey, E.V.: Effects of previous antiresorptive
therapy on the bone mineral density response to two years of teriparatide treatment in
postmenopausal women with osteoporosis. J. Clin. Endocrinol. Metab. 93, 852–860 (2008)
118. Miller, P.D., Delmas, P.D., Lindsay, R., Watts, N.B., Luckey, M., Adachi, J., Saag, K.,
Greenspan, S.L., Seeman, E., Boonen, S., Meeves, S., Lang, T.F., Bilezikian, J.P.: Early
responsiveness of women with osteoporosis to teriparatide after therapy with alendronate or
risedronate. J. Clin. Endocrinol. Metab. 93, 3785–3793 (2008)
119. Stepan, J.J., Burr, D.B., Li, J., Ma, Y.L., Petto, H., Sipos, A., Dobnig, H., Fahrleitner-
Pammer, A., Michalska, D., Pavo, I.: Histomorphometric changes by teriparatide in
alendronate-pretreated women with osteoporosis. Osteoporos. Int. 5, 5 (2010)
120. Black, D.M., Greenspan, S.L., Ensrud, K.E., Palermo, L., McGowan, J.A., Lang, T.F.,
Garnero, P., Bouxsein, M.L., Bilezikian, J.P., Rosen, C.J.: The effects of parathyroid
hormone and alendronate alone or in combination in postmenopausal osteoporosis. N. Engl.
J. Med. 349, 1207–1215 (2003)
121. Greenspan, S.L., Emkey, R.D., Bone, H.G., Weiss, S.R., Bell, N.H., Downs, R.W., McKeever, C.,
Miller, S.S., Davidson, M., Bolognese, M.A., Mulloy, A.L., Heyden, N., Wu, M., Kaur, A.,
Lombardi, A.: Significant differential effects of alendronate, estrogen, or combination therapy on
the rate of bone loss after discontinuation of treatment of postmenopausal osteoporosis. A
randomized, double-blind, placebo-controlled trial. Ann. Intern. Med. 137, 875–883 (2002)
122. Cosman, F., Wermers, R.A., Recknor, C., Mauck, K.F., Xie, L., Glass, E.V., Krege, J.H.:
Effects of teriparatide in postmenopausal women with osteoporosis on prior alendronate or
raloxifene: differences between stopping and continuing the antiresorptive agent. J. Clin.
123. Vestergaard, P., Jorgensen, N.R., Mosekilde, L., Schwarz, P.: Effects of parathyroid
hormone alone or in combination with antiresorptive therapy on bone mineral density and
fracture risk–a meta-analysis. Osteoporos. Int. 18, 45–57 (2007)
124. Chevalier, Y., Quek, E., Borah, B., Gross, G., Stewart, J., Lang, T., Zysset, P.:
Biomechanical effects of teriparatide in women with osteoporosis treated previously with
alendronate and risedronate: results from quantitative computed tomography-based finite
element analysis of the vertebral body. Bone 46, 41–48 (2010)
125. Ma, Y.L., Bryant, H.U., Zeng, Q., Schmidt, A., Hoover, J., Cole, H.W., Yao, W., Jee, W.S.,
Sato, M.: New bone formation with teriparatide [human parathyroid hormone-(1-34)] is not
retarded by long-term pretreatment with alendronate, estrogen, or raloxifene in
ovariectomized rats. Endocrinology 144, 2008–2015 (2003)
126. Samadfam, R., Xia, Q., Goltzman, D.: Pretreatment with anticatabolic agents blunts but
does not eliminate the skeletal anabolic response to parathyroid hormone in
oophorectomized mice. Endocrinology 148, 2778–2787 (2007)
127. Yao, W., Su, M., Zhang, Q., Tian, X., Setterberg, R.B., Blanton, C., Lundy, M.W., Phipps, R.,
Jee, W.S.: Risedronate did not block the maximal anabolic effect of PTH in aged rats. Bone 41,
813–819 (2007)
128. Minne, H., Audran, M., Simoes, M.E., Obermayer-Pietsch, B., Sigurethsson, G., Marin, F.,
Dalsky, G.P., Group, T.N.: Bone density after teriparatide in patients with or without prior
antiresorptive treatment: one-year results from the EUROFORS study. Curr. Med. Res.
Opin. 4, 4 (2008)
129. Obermayer-Pietsch, B.M., Marin, F., McCloskey, E.V., Hadji, P., Farrerons, J., Boonen, S.,
Audran, M., Barker, C., Anastasilakis, A.D., Fraser, W.D., Nickelsen, T.: Effects of two
years of daily teriparatide treatment on BMD in postmenopausal women with severe
osteoporosis with and without prior antiresorptive treatment. J. Bone Miner. Res. 23, 1591–
1600 (2008)
130. Deal, C., Omizo, M., Schwartz, E.N., Eriksen, E.F., Cantor, P., Wang, J., Glass, E.V.,
Myers, S.L., Krege, J.H.: Combination teriparatide and raloxifene therapy for
postmenopausal osteoporosis: results from a 6-month double-blind placebo-controlled
trial. J. Bone Miner. Res. 20, 1905–1911 (2005)
131. Black, D.M., Bilezikian, J.P., Ensrud, K.E., Greenspan, S.L., Palermo, L., Hue, T., Lang, T.F.,
McGowan, J.A., Rosen, C.J.: One year of alendronate after one year of parathyroid hormone
(1-84) for osteoporosis. N. Engl. J. Med. 353, 555–565 (2005)
132. Vahle, J.L., Sato, M., Long, G.G., Young, J.K., Francis, P.C., Engelhardt, J.A., Westmore, M.S.,
Linda, Y., Nold, J.B.: Skeletal changes in rats given daily subcutaneous injections of
recombinant human parathyroid hormone(1-34) for 2 years and relevance to human safety.
Toxicol. Pathol. 30, 312–321 (2002)
133. Lindsay, R., Scheele, W.H., Neer, R., Pohl, G., Adami, S., Mautalen, C., Reginster, J.Y.,
Stepan, J.J., Myers, S.L., Mitlak, B.H.: Sustained vertebral fracture risk reduction after
withdrawal of teriparatide in postmenopausal women with osteoporosis. Arch. Intern. Med.
164, 2024–2030 (2004)
134. Black, D.M., Cummings, S.R., Karpf, D.B., Cauley, J.A., Thompson, D.E., Nevitt, M.C.,
Bauer, D.C., Genant, H.K., Haskell, W.L., Marcus, R., Ott, S.M., Torner, J.C., Quandt, S.A.,
Reiss, T.F., Ensrud, K.E.: Randomized trial of effect of alendronate on risk of fracture in
women with existing vertebral fractures. Fracture intervention trial research group. Lancet
348, 1535–1541 (1996)
135. Pols, H.A., Felsenberg, D., Hanley, D.A., Stepan, J., Munoz-Torres, M., Wilkin, T.J.,
Qin-sheng, G., Galich, A.M., Vandormael, K., Yates, A.J., Stych, B.: Multinational,
placebo-controlled, randomized trial of the effects of alendronate on bone density and
fracture risk in postmenopausal women with low bone mass: results of the FOSIT study.
Fosamax International trial study group. Osteoporos. Int. 9, 461–468 (1999)
136. Harris, S.T., Watts, N.B., Genant, H.K., McKeever, C.D., Hangartner, T., Keller, M.,
Chesnut 3rd, C.H., Brown, J., Eriksen, E.F., Hoseyni, M.S., Axelrod, D.W., Miller, P.D.:
Effects of risedronate treatment on vertebral and nonvertebral fractures in women with
postmenopausal osteoporosis: a randomized controlled trial. Vertebral efficacy with
risedronate therapy (VERT) study group. JAMA 282, 1344–1352 (1999)
137. Chesnut, I.C., Skag, A., Christiansen, C., Recker, R., Stakkestad, J.A., Hoiseth, A.,
Felsenberg, D., Huss, H., Gilbride, J., Schimmer, R.C., Delmas, P.D.: Effects of oral
ibandronate administered daily or intermittently on fracture risk in postmenopausal
osteoporosis. J. Bone Miner. Res. 19, 1241–1249 (2004)
138. Black, D.M., Delmas, P.D., Eastell, R., Reid, I.R., Boonen, S., Cauley, J.A., Cosman, F.,
Lakatos, P., Leung, P.C., Man, Z., Mautalen, C., Mesenbrink, P., Hu, H., Caminis, J., Tong, K.,
Rosario-Jansen, T., Krasnow, J., Hue, T.F., Sellmeyer, D., Eriksen, E.F., Cummings, S.R.:
Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N. Engl. J. Med.
356, 1809–1822 (2007)
139. Jansen, J.P., Bergman, G.J., Huels, J., Olson, M.: Prevention of vertebral fractures in
osteoporosis: mixed treatment comparison of bisphosphonate therapies. Curr. Med. Res.
Opin. 25, 1861–1868 (2009)
Bone Cell Mechanoresponsiveness
Damian C. Genetos and Christopher R. Jacobs
Abstract Improvements in health and nutrition have increased human longevity

over the past centuries. Beneficial increases in lifespan, however, are met with novel
degenerative and age-related diseases, including atherosclerosis and heart disease,
sarcopenia, diabetes, and osteoporosis. The effects of aging upon the skeleton include
anatomically heterogeneous involution, alterations in the composition of both the
organic and mineral component of the matrix, and accumulation of microdamage.
It has been suggested that the process of mechanotransduction within bone is also
affected by aging. Within this chapter, we review the process of mechanotransduc-
tion in osteogenic cells, highlight those works that have examined age-related
changes in mechanotransduction, and discuss mechanosensitive systems implicated
in other tissues.
1 Introduction
External mechanical forces originating from the environment exert critical roles in
development and homeostasis. For example, during embryonic development,
intracardiac shear forces are required for cardiac morphogenesis [36]. Similarly,
D. C. Genetos (&)
Department of Anatomy, Physiology, and Cell Biology,
UC Davis School of Veterinary Medicine, Davis,
CA 95616, USA
e-mail: dgenetos@ucdavis.edu
C. R. Jacobs
Department of Biomedical Engineering, Columbia University,
New York, NY 10027, USA

DOI: 10.1007/8415_2011_109
Published Online: 14 October 2011
178 D. C. Genetos and C. R. Jacobs
left–right patterning of the primary visceral organs is caused by ciliary left-ward

fluid flow in Hensen’s node [18, 39]. In the adult, resistance exercise promotes
skeletal muscle hypertrophy to functionally adapt to increases in applied loads
[33]. Just as mechanical forces promote functional adaptations to a changing
environment, perturbations in sensing and responding to mechanical forces play a
role in such diseases as cancer, atherosclerosis, asthma, and osteoporosis. Muta-
tions in dynein, a motor protein associated with cilia, can prevent left-ward fluid
flow in Hensen’s node during embryogenesis; this causes the left–right patterning
of the internal organs to be random (heterotaxy) or reversed (situs inversus)
[39, 53, 93, 95]. Within the cardiovascular system, laminar hemodynamic forces in
the heart are thought to be anti-atherogenic, whereas disturbed, turbulent flow is
thought to promote vascular remodeling and atherogenesis [9, 73, 89, 109]. In the
absence of sufficient load, skeletal muscle undergoes atrophy.
The mammalian skeleton demonstrates tremendous capacity for functional
adaptation to mechanical forces. For example, conditions of reduced skeletal stress
promote bone resorption to minimize unnecessary energy expenditures, while
increased skeletal stress promotes bone formation that increases bone mass and
thereby reduces stress and/or strain upon subsequent loading. At the tissue and
cellular level, external mechanical forces promote changes in skeletal architecture by
altering proliferation and self-renewal, differentiation, matrix production and min-
eralization. Interestingly, the anabolic effect of physical activity promotes larger
changes in bone mass and strength in the young compared to the elderly [6, 41].
A variety of factors—decreased muscle mass to generate load on bone [29], dimin-
ished sensitivity to mechanical stimuli with increased age [84, 99, 100], decreased
osteocyte number with age [26, 62], or diminished capacity for matrix formation
[69, 98, 104]—may be altered, however, the relative roles of each are unknown.
Interestingly, there are conflicting reports as to the effects that aging has upon me-
chanoresponsiveness in vivo—for example, Rubin et al. showed that loading of
isolated turkey ulnae produced periosteal bone formation that was attenuated in older
animals [84, 100] demonstrated a change in the threshold of mechanical strain
required for bone formation in young (9 month) versus old (19 month) rats. In con-
trast, Brodt and Silva recently demonstrated no loss of responsiveness to tibial
compression in aged mice compared to young adults [13]. Furthermore, exercise
studies in rodents show either reduced responsiveness with age [92, 35], no influence
of age [40, 77, 102], and even enhanced responsiveness with age [15, 54]. Despite the
appreciation of the socioeconomic burden of osteoporosis, the field of age-related
changes in mechanotransduction remains surprisingly fallow.
2 Mechanotransduction in Bone Cells
Mechanotransduction refers to the series of inter-related processes wherein

mechanical loading of the skeleton produces an adaptive response. All cells and
organisms are responsive to mechanical forces [38, 39, 68] however, with the
Bone Cell Mechanoresponsiveness 179
Fig. 1 Overview of Mechanical loading

processes and players in
skeletal mechanotransduction ↓
Localized biophysical signal

substrate strain, fluid shear stress
Mechanosensitive bone cells

MSCs, osteoblasts, osteocytes
Biochemical responses and

signaling pathways
Ca 2+i , ATP, NO, PG, β-catenin
Skeletal adaptation
Osteoblasts and osteoclasts
exception of specialized excitable cells involved in hearing and touch, the cellular
mechanisms responsible for conversion of an external force (e.g., gravity, strain,
hydrostatic pressure, fluid shear stress) into a cellular response generally remain
poorly understood. Burger et al. [16] described the processes of bone or skeletal
mechanotransduction to involve (Fig. 1):
1. Initiation of a cell-level biophysical signal that can be perceived by an
osteocyte;
2. Transduction of this biophysical signal into a biochemical signal;
3. Communication of this biochemical signal to effector cells (osteoblasts or
osteoclasts).
A variety of mechanisms have been proposed that convert an external load into
a pericellular biophysical signal within bone, be it an osteocyte, osteoblast, bone-
lining cell, or mesenchymal stem cell within the bone marrow stroma; such signals
include piezoelectric fields, substrate strain due to deformation of the ECM,
hydrostatic pressure, and, under certain conditions, bone tissue damage (for greater
detail, the reader is directed to the works of Duncan and Turner [25], Robling et al.
[82], Turner et al. [101], or Rubin et al. [85], amongst many others). Due to its
ability to elicit cellular responses at physiological loading levels in in vitro studies,
the current paradigm is that shear stress due to fluid flow across the body or
dendritic processes of mechanosensory cells is the most likely mechanism
whereby mechanical loading, such as occur during high-impact exercise, elicit
functional skeletal adaptation [70, 107]. It should be added that direct substrate
strain, which has been excluded by many investigators due to the absence or
relative paucity of in vitro cellular response to in vivo levels of physiologic

strain, may have been underestimated as a biophysical signal. There is increasing
evidence for localized amplification of substrate strain, such that tissue-level
strains may be amplified to the point where they can induce a cellular response
[10, 64, 78, 108, 111].
Biophysical signals activate various signaling pathways resulting in both pro-
osteogenic and anti-catabolic outcomes. Rapid responses (0 s–1 min) to such
signals include the generation or liberation of second messengers like Ca2+, cAMP,
DAG, and IP3. These, in turn, promote the synthesis and secretion of autocrine
paracrine factors (e.g., NO, PGs, and IGFs), kinase activation, cytoskeletal rear-
rangement, transcription factor (NF-jB, b-catenin, ATF4) activity, and changes in
gene transcription and translation. These biochemical signals are communicated
(via gap junctions, integrins, and soluble factors) to effector cells that are
responsible for initiation of tissue-level responses. In the context of skeletal
homeostasis or adaptation, this is the concerted activity of the bone multicellular
unit (BMU) for remodeling, or, in situations of modeling, bone deposition by
osteoblasts or resorption by osteoclasts.
3 Does Aging Influence Mechanotransduction in Bone Cells?
While the pathogenesis of osteoporosis and senescence-related bone loss is mul-

tifactorial, it is thought than one factor may be reduced bone cell mechanosensi-
tivity and/or mechanoresponsiveness [63]. Surprisingly few studies have directly
examined whether the ability of osteoblastic cells to perceive or to respond to
mechanical stimuli changes with age and, if so, whether that is due to altered
mechanosensitivity or mechanoresponsiveness.
One of the earliest responses of osteoblastic and osteocytic cells to fluid shear
stress is an increase in intracellular calcium levels (Ca2+i) which may require either
release from intracellular stores [60, 110, 112], entry through ion channels from
extracellular fluid [79, 87], or both [37]. This change in Ca2+i is observed in MSCs,
osteoblasts, and osteocytes after onset of shear stress. However, there is evidence
that these cells are not equally responsive to a given stimulus: Kamioka et al.
demonstrated that shear stresses of 1.2 or 2.4 Pa (n.b., 1 Pa = 10 dynes/cm2) were
less stimulatory to primary osteocytes, in terms of percent of cells responding with
an increase in Ca2+i, than were primary osteoblasts [44]. Despite differences in
sensitivity to applied stresses, those osteocytes which did respond demonstrated no
significant change in the magnitude of the Ca2+i response, suggesting that the
difference is in the ability of the cells to perceive, but not respond to, shear stress.
This assumes that the magnitude of the Ca2+i response is most important; some
evidence suggests that other parameters, such as the frequency of applied load, is
more important [8, 23].
To date, the only study designed to examine whether the Ca2+i response of
osteoblastic cells to fluid shear stress was a function of donor age was performed
by Donahue et al. [24]. They observed a modest, but statistically significant,

reduction in the frequency of Ca2+i oscillations under both static and fluid flow
conditions in osteoblasts derived from old rats (24 mos.) versus osteoblasts derived
from mature rats (12 mos.); interestingly, there was no difference in the magnitude
of the response across the age groups, similar to the study of Kamioka et al.
Increases in Ca2+i signaling are linked to the expression of markers implicated
in bone modeling and remodeling, such as insulin-like growth factor (IGF)
[47, 61], nitric oxide (NO) [42, 59, 79], prostaglandins [42, 47, 79], TGF-b1 [88],
and ATP [30]. While Donahue et al. remains the only study to examine whether
the Ca2+i response to fluid shear stress is affected by donor age, a larger number of
studies have reported whether the release of such paracrine and autocrine factors as
NO, IGF, and prostaglandins, is affected by osteoporosis.
Sterck et al. [94] examined whether osteoblastic cells derived from trabecular
bone biopsies normal (mean age 67 y) or osteoporotic donors (mean age 61 y)
were similarly affected by mechanical stress. Cells from both donor groups
demonstrated an increase in alkaline phosphatase activity and osteocalcin release
in response to 10 nM 1a, 25-dihydroxyvitamin D3, indicating that both cell sources
were osteoblastic in phenotype. In response to a pulsatile shear stress of
0.7 ± 0.03 Pa applied at a frequency of 5 Hz, osteoblastic cells from both non-
osteoporotic and osteoporotic donors significantly increased the release of pros-
taglandin E2 (PGE2) and NO after 1 h. A difference between non-osteoporotic and
osteoporotic donor cell responsiveness was only found 24 h after cessation of fluid
shear stress, at which point cells from non-osteoporotic donors continued to release
PGE2, whereas cells from osteoporotic donors did not. Since PGE2 and NO release
are influenced by fluid shear stress-induced Ca2+i oscillations, these data suggest
that the modest decrease in the percent of aged osteoblastic cells to respond to fluid
flow, as observed by Donahue et al. [24], may not affect the ability of an osteo-
blastic cell to generate of soluble mediators like NO or PGE2.
In a related study, Bakker et al. examined the influence of pulsatile fluid shear
stress upon NO and PGE2 release in primary bone cells cultured from osteoporotic
(OP) or osteoarthritic (OA) donors [5]. Osteoblastic cells from both OA and OP
donors responded to a pulsatile shear stress of 0.6 ± 0.3 Pa (5 Hz) by increasing
release of NO and PGE2; increased PGE2 release was paralleled with increased
transcription of COX-2, the inducible enzyme responsible for PG synthesis. This
would indicate that osteoblastic bone cells from aged (mean age 78 y for OP, 75 y
for OA), diseased individuals are still capable of responding to an in vitro
mechanical stimulus, suggesting that the capacity of cells to respond is unaffected
by aging or disease state. They also examined whether mechanosensitivity was
different in OA versus OP cultures, by examining the NO and PGE2 response at
varying shear rates. Varying the magnitude of applied shear stress revealed dif-
ferences in mechanoresponsiveness between the OA and OP groups. Both PGE2
and NO release were greater in OP-derived than OA-derived osteoblasts at lower
shear stress (0.4 ± 0.1 Pa). However, since there was no comparison to osteo-
blastic cells from asymptomatic elderly patients, one cannot conclude whether this
differential mechanosensitivity is due to disease. The same group did examine the
influence of donor cell age in an earlier study [48], wherein they reported that
pulsatile shear stress-induced PGE2 and PGI2 release increased with donor
cell age.
In summary, these limited data suggest that there is little intrinsic influence of
aging or disease state upon the ability of in vitro osteoblastic cells to perceive or
respond to mechanical stresses. Whether the same occurs in vivo is unknown, but
this instead suggests that additional factors, which influence bone cell function and
which change with aging, may exert potent influences upon load-induced bone
formation. Nonetheless, there remains doubt as to whether these questions have
been studied sufficiently; perhaps the effect of age is on the mechanical regulation
of mesenchymal stem cells or osteoprogenitors, a topic even less studied.
4 Sex Steroids and Mechanotransduction
The tissue-level and intracellular processes involved in mechanotransduction are

influenced not only by local factors produced during the conversion of external
mechanical forces into localized cellular responses, but also by humoral factors
within the circulation. Estrogen has long been postulated as a pivotal mediator of
bone cell mechanoresponsiveness. Frost proposed that estrogen may reduce the
minimum effective strain required for an anabolic response to load. This would
enable strains magnitudes that were previously sub-threshold and therefore
insufficient to cause bone formation, to initiate a bone-forming response [28].
Estrogen influences skeletal homeostasis through such avenues as progenitor
recruitment, proliferation, differentiation and apoptosis [80]. Thus, the loss of
estrogen production, as occurs during menopause, is thought to be a causative
factor in age-associated bone loss, and this is supported by bone loss in ovariec-
tomized mice [14]. Yet, whereas ample in vivo data support the hypothesis that the
loss of estrogen influences skeletal homeostasis and adaptation, there is, again,
little in vitro evidence for an effect upon mechanotransduction. Whereas certain
studies, described below, have examined the influence of exogenous estrogen upon
the response of bone cells to shear stresses in vitro, this does not address the in
vivo phenomenon wherein estrogen is no longer produced.
Mechanistically, estrogen (as well as progesterone and testosterone, other ste-
roids) initiate cellular responses via two general mechanisms, the classical genomic
pathway, or the non-genomic pathway. Canonical estrogen signaling involves ligand
diffusion across the plasmalemma to the nucleus, binding to estrogen receptors, and
induction of gene transcription [1]. Over the past decade, accumulating evidence
indicates a role for rapid, non-genomic action of estrogen, the effects of which
include microtubule polymerization, second messenger formation, and kinase acti-
vation (for elaboration, the reader is directed to such works as [50, 58, 106]).
Early ex vivo evidence of a role for estrogen in mechanostransduction came
from the work of Lanyon and associates. Using ulnar explants from rats, they
observed a synergistic influence of 10 nM estrogen and cyclical loading (-1300 le
on medial side to 700 le on lateral side at 1 Hz for 8 min) upon [3H]thymidine

(index of proliferation) and [3H]proline (matrix formation) incorporation com-
pared to estrogen or load alone [20]. Extensions of this work have helped to
elucidate the role of estrogens and the estrogen receptors in mediating the adaptive
response to load (c.f., [2, 19, 27, 50, 51, 67, 75]).
Because osteogenic cells appear to retain a memory of their donor environment
[49, 86], one would expect osteoblastic cells derived from post-menopausal sub-
jects to demonstrate a differential mechanoresponse compared to cells derived
from pre-menopausal subjects. While such a direct comparison has not been made,
investigators have demonstrated that estrogen augments the response to mechan-
ical load. Joldersma et al. revealed that two days of culture in 10 pM estrogen
significantly enhanced PGE2 release from osteoblastic cells from elderly woman
(age 56–75 years, non-osteoporotic) compared to cells exposed to pulsatile fluid
flow without estrogen pre-treatment [43]. Similarly, Bakker et al. demonstrated
that 24 h of 10 pM estrogen provided a synergistic response to pulsatile fluid
flow in osteoblastic cells from osteoporotic women (mean age 82; range
62–90 years) [4].
What these studies do not show, unfortunately, are age-related decreases in a
biological response to loading that can be rescued by estrogen treatment. Nor do
they address a mechanism whereby estrogen exerts an additive effect upon bone
cell mechanoresponsiveness. Interestingly, Armstrong et al. showed that the
estrogen receptors (ER) plays an obligate role in Wnt signaling [3], as ER
antagonism with ICI 182,780 or tamoxifen prevented nuclear accumulation of
b-catenin. While not demonstrated to date, one would predict decreased
Wnt signaling in cells derived from post-menopausal subjects compared to pre-
menopausal subjects. The importance of this pathway, through both canonical and
non-canonical mechanisms, in skeletal development and adaptation to load is well-
documented [31, 45, 56, 57, 66, 81, 90, 91].
Aguirre et al. have provided a model for ligand-independent ER function in
mediating mechanotransduction [2]. They observed attenuated strain-induced
ERK1/2 phosphorylation in cells derived from ERa-/- or ERb-/- mice.
Remarkably, transfection with the ligand-binding domain of either receptor
restored the ability of these cells to increase ERK1/2 phosphorylation in response
to strain, as did transfection with an ERa mutant that is unable to bind estrogens.
The role of ERa and ERb in mediating these processes appears to involve non-
genomic signaling, based upon the rapid time of response (10 min) and because
mechanoresponsiveness was restored using a plasmalemma-targeted, but not
nuclear-targeted, ER. These findings would suggest, therefore, that bone cell
mechanoresponsiveness would not be affected by estrogen status, if ER expression
does not change with aging. However, there is evidence that ER expression also
decreases with age [7, 11, 12, 34], perhaps reducing the potential for mechano-
transduction via any estrogen-independent ER mechanism.
5 Humoral and Paracrine Factors in Mechanotransduction
Just as estrogen plays a permissive or synergistic role in bone cell mechano-

transduction, so also do humoral and paracrine factors; two of these, parathyroid
hormone (PTH) and insulin-like growth factors (IGFs), reveal age-related effects
in bone cell mechanoresponsiveness.
PTH plays diverse roles in skeletal adaptation. Intermittent doses of PTH(1–34)
stimulate bone formation [32], whereas continuous administration promotes bone
resorption [71]. PTH treatment synergizes with mechanical loading to enhance
bone formation [55] and is required for load-induced bone formation, perhaps by
sensitizing either the strain-sensing mechanism itself or early responses of bone to
strain-generated signals [21]. Donahue et al. have shown that osteoblastic cells
from aged rodents (24–28 mos.) demonstrate an impaired cAMP response to PTH
compared to osteoblastic cells isolated from young (4 mo.) rats, suggesting an age-
related decrease in G protein signaling in osteoblastic cells [22].
IGFs are anabolic agents within bone, where they function to increase cell
number and decrease apoptosis in osteoprogenitors, and stimulate osteoblast
recruitment to the bone surface [46]. Like PTH, IGF synergizes with mechanical
loading [97], and bone cell responsiveness to IGF decreases with age [52, 83]
through decreased IGF-IR signaling [17].
6 Attenuated Response to Loading in the Aged May be Due

to Fewer Mechanosensory Cells
All osteogenic cell types within the skeleton—the MSC, the osteoblast, the bone-
lining cell, and the osteocyte—are mechanoresponsive. Because the evidence
examined above does not suggest a diminished capacity of these cells to respond to
load, one can consider the alternate hypothesis that there are simply fewer of these
cells present within the skeleton to respond to load. MSC frequency within the
bone marrow compartment is on the order of 1 per 105 stromal cell in younger
individuals [74], and this declines with age [65, 76], in part due to reduced
plasticity of MSCs and a concomitant shift toward an adipocytic phenotype [105].
Such reductions in MSC frequency, and general phenotypic drift towards an
adipocyte, would decrease the number of osteoblasts capable of perceiving
mechanical loads on their own, or biochemical signals from responding osteocytes.
Osteocytes are considered the most important mechanosensor within bone: they
are present uniformly throughout bone, they are the most abundant cell type within
bone (ten fold greater number than osteoblasts [72]), and their dendritic processes
allow for direct cell–cell communication to neighboring osteocytes, osteoblasts,
and mesenchymal stem cells. Indeed, osteocyte ablation prevented disuse-induced
bone loss in a tail-limb suspension model [96].1 As with MSCs, osteocyte number
decreases with age [62], and, quite interestingly, osteocyte number is different in
males and females [103].
7 Conclusions
There is in vivo evidence both for, and against, diminished skeletal adaptation to
load with age. If in fact there is diminished mechanotransduction with age, there
appears to be little evidence that this is due to intrinsic deficits in the ability of an
osteogenic cell to respond to loading. While aged cells demonstrate decreased
responsiveness to agents like PTH or IGF, the limited available data suggest that
there is little intrinsic influence of aging or disease state upon the ability of in vitro
osteoblastic cells to perceive or respond to mechanical stresses.
Acknowledgments This work was supported by NIH NIAMS R03 AR57547 (DCG) and NIH
NIAMS R01 AR45989, NIAMS R21 AR45156, and New York Stem Cell Grant N06G-210
(CRJ). The authors are grateful to Dr. C.E. Yellowley for suggestions to Fig. 1, and to Dr. R.Y.
Kwon for helpful discussion.
References
1. Acconcia, F., Kumar, R.: Signaling regulation of genomic and nongenomic functions of
estrogen receptors. Cancer Lett. 238(1), 1–14 (2006)
2. Aguirre, J.I., Plotkin, L.I., et al.: A novel ligand-independent function of the estrogen
receptor is essential for osteocyte and osteoblast mechanotransduction. J. Biol. Chem.
282(35), 25501–25508 (2007)
3. Armstrong, V.J., Muzylak, M., et al.: Wnt/beta-catenin signaling is a component of
osteoblastic bone cell early responses to load-bearing and requires estrogen receptor alpha.
J. Biol. Chem. 282(28), 20715–20727 (2007)
4. Bakker, A.D., Klein-Nulend, J., et al.: Additive effects of estrogen and mechanical stress on
nitric oxide and prostaglandin E2 production by bone cells from osteoporotic donors.
Osteoporos. Int. 16(8), 983–989 (2005)
5. Bakker, A.D., Klein-Nulend, J., et al.: Different responsiveness to mechanical stress of bone
cells from osteoporotic versus osteoarthritic donors. Osteoporos. Int. 17(6), 827–833 (2006)
6. Bassey, E.J., Rothwell, M.C., et al.: Pre- and postmenopausal women have different bone
mineral density responses to the same high-impact exercise. J. Bone Miner. Res. 13(12),
1805–1813 (1998)
1
Interestingly, transgenic mice that lost bone under tail suspension did not require osteocytes to
regain bone mass, suggesting that osteocytes are indispensible for bone loss due to unloading,
whereas reloading-induced increases in bone mass are osteocyte-independent. These data indicate
that, in the absence of osteocytes, osteoblasts are sufficiently mechanosensitive and mechano-
responsive to initiate skeletal adaptation.
7. Batra, G.S., Hainey, L., et al.: Evidence for cell-specific changes with age in expression of
oestrogen receptor (ER) alpha and beta in bone fractures from men and women. J. Pathol.
200(1), 65–73 (2003)
8. Batra, N.N., Li, Y.J., et al.: Effects of short-term recovery periods on fluid-induced signaling
in osteoblastic cells. J. Biomech. 38(9), 1909–1917 (2005)
9. Berk, B.C., Abe, J.I., et al.: Endothelial atheroprotective and anti-inflammatory
mechanisms. Ann. N. Y. Acad. Sci. 947, 93–109 (2001). Discussion 109–111
10. Bonivtch, A.R., Bonewald, L.F., et al.: Tissue strain amplification at the osteocyte lacuna: a
microstructural finite element analysis. J. Biomech. 40(10), 2199–2206 (2007)
11. Braidman, I., Baris, C., et al.: Preliminary evidence for impaired estrogen receptor-alpha
protein expression in osteoblasts and osteocytes from men with idiopathic osteoporosis.
Bone 26(5), 423–427 (2000)
12. Braidman, I.P., Baris, C., et al.: Preliminary report of impaired oestrogen receptor-alpha
expression in bone, but no involvement of androgen receptor, in male idiopathic osteoporosis.
J. Pathol. 192(1), 90–96 (2000)
13. Brodt, M.D., Silva, M.J.: Aged mice have enhanced endocortical response and normal
periosteal response compared with young-adult mice following 1 week of axial tibial
compression. J. Bone Miner. Res. 25(9), 2006–2015 (2010)
14. Broulik, P.D.: Tamoxifen prevents bone loss in ovariectomized mice. Endocr. Regul. 25(4),
217–219 (1991)
15. Buhl, K.M., Jacobs, C.R., et al.: Aged bone displays an increased responsiveness to
low-intensity resistance exercise. J. Appl. Physiol. 90(4), 1359–1364 (2001)
16. Burger, E.H., Klein-Nulend, J.: Mechanotransduction in bone–role of the lacuno-canalicular
network. Faseb J. 13, S101–S112 (1999)
17. Cao, J.J., Kurimoto, P., et al.: Aging impairs IGF-I receptor activation and induces skeletal
resistance to IGF-I. J. Bone Miner. Res. 22(8), 1271–1279 (2007)
18. Carlen, B., Stenram, U.: Primary ciliary dyskinesia: a review. Ultrastruct. Pathol. 29(3–4),
217–220 (2005)
19. Chen, J.R., Plotkin, L.I., et al.: Transient versus sustained phosphorylation and nuclear
accumulation of ERKs underlie anti-versus pro-apoptotic effects of estrogens. J. Biol.
Chem. 280(6), 4632–4638 (2005)
20. Cheng, M.Z., Zaman, G., et al.: Mechanical loading and sex hormone interactions in organ
cultures of rat ulna. J. Bone Miner. Res. 11(4), 502–511 (1996)
21. Chow, J.W., Fox, S., et al.: Role for parathyroid hormone in mechanical responsiveness of
rat bone. Am. J. Physiol. 274(1 Pt 1), E146–E154 (1998)
22. Donahue, H.J., Zhou, Z., et al.: Age-related decreases in stimulatory G protein-coupled
adenylate cyclase activity in osteoblastic cells. Am. J. Physiol. 273(4 pt 1), E776–E781 (1997)
23. Donahue, S.W., Donahue, H.J., et al.: Osteoblastic cells have refractory periods for fluid-
flow-induced intracellular calcium oscillations for short bouts of flow and display multiple
low-magnitude oscillations during long-term flow. J. Biomech. 36(1), 35–43 (2003)
24. Donahue, S.W., Jacobs, C.R., et al.: Flow-induced calcium oscillations in rat osteoblasts are
age, loading frequency, and shear stress dependent. Am. J. Physiol. Cell Physiol. 281(5),
C1635–C1641 (2001)
25. Duncan, R.L., Turner, C.H.: Mechanotransduction and the functional response of bone to
mechanical strain. Calcif. Tissue Int. 57, 344–358 (1995)
26. Dunstan, C.R., Somers, N.M., et al.: Osteocyte death and hip fracture. Calcif. Tissue Int.
53(1), S113–S116 (1993). Discussion S116–S117
27. Emerton, K.B., Hu, B., et al.: Osteocyte apoptosis and control of bone resorption following
ovariectomy in mice. Bone 46(3), 577–583 (2010)
28. Frost, H.M.: Vital biomechanics: proposed general concepts for skeletal adaptations to
mechanical usage. Calcif. Tissue Int. 42(3), 145–156 (1988)
29. Frost, H.M.: Why do bone strength and ‘‘mass’’ in aging adults become unresponsive to
vigorous exercise? Insights of the Utah paradigm. J. Bone Miner. Metab. 17(2), 90–97
(1999)
30. Genetos, D.C., Geist, D.J., et al.: Fluid shear-induced ATP secretion mediates prostaglandin
release in MC3T3–E1 osteoblasts. J. Bone Miner. Res. 20(1), 41–49 (2005)
31. Gong, Y., Slee, R.B., et al.: LDL receptor-related protein 5 (LRP5) affects bone accrual and
eye development. Cell 107(4), 513–523 (2001)
32. Gunness-Hey, M., Hock, J.M.: Increased trabecular bone mass in rats treated with human
synthetic parathyroid hormone. Metab. Bone Dis. Relat. Res. 5(4), 177–181 (1984)
33. Haberland, M., Montgomery, R.L., et al.: The many roles of histone deacetylases in
development and physiology: implications for disease and therapy. Nat. Rev. Genet. 10(1),
32–42 (2009)
34. Han, Z., Kokkonen, G.C., et al.: Effect of aging on populations of estrogen receptor-
containing cells in the rat uterus. Exp. Cell Res. 180(1), 234–242 (1989)
35. Hoshi, A., Watanabe, H., et al.: Effects of exercise at different ages on bone density and
mechanical properties of femoral bone of aged mice. Tohoku J. Exp. Med. 185(1), 15–24
(1998)
36. Hove, J.R., Koster, R.W., et al.: Intracardiac fluid forces are an essential epigenetic factor
for embryonic cardiogenesis. Nature 421(6919), 172–177 (2003)
37. Hung, C.T., Allen, F.D., et al.: Intracellular Ca2+ stores and extracellular Ca2+ are required
in the real-time Ca2+ response of bone cells experiencing fluid flow. J. Biomech. 29(11),
1411–1417 (1996)
38. Ingber, D.E.: Cellular mechanotransduction: putting all the pieces together again. FASEB J.
20(7), 811–827 (2006)
39. Jaalouk, D.E., Lammerding, J.: Mechanotransduction gone awry. Nat. Rev. Mol. Cell Biol.
10(1), 63–73 (2009)
40. Jarvinen, T.L., Pajamaki, I., et al.: Femoral neck response to exercise and subsequent
deconditioning in young and adult rats. J. Bone Miner. Res. 18(7), 1292–1299 (2003)
41. Jee, W.S., Tian, X.Y.: The benefit of combining non-mechanical agents with mechanical
loading: a perspective based on the Utah Paradigm of Skeletal Physiology. J. Musculoskelet.
Neuronal Interact. 5(2), 110–118 (2005)
42. Jessop, H.L., Rawlinson, S.C., et al.: Mechanical strain and fluid movement both activate
extracellular regulated kinase (ERK) in osteoblast-like cells but via different signaling
pathways. Bone 31(1), 186–194 (2002)
43. Joldersma, M., Klein-Nulend, J., et al.: Estrogen enhances mechanical stress-induced
prostaglandin production by bone cells from elderly women. Am. J. Physiol. Endocrinol.
Metab. 280(3), E436–E442 (2001)
44. Kamioka, H., Sugawara, Y., et al.: Fluid shear stress induces less calcium response in a
single primary osteocyte than in a single osteoblast: implication of different focal adhesion
formation. J. Bone Miner. Res. 21(7), 1012–1021 (2006)
45. Kato, M., Patel, M.S., et al.: Cbfa1-independent decrease in osteoblast proliferation,
osteopenia, and persistent embryonic eye vascularization in mice deficient in Lrp5, a Wnt
coreceptor. J. Cell Biol. 157(2), 303–314 (2002)
46. Kawai, M., Rosen, C.J.: Insulin-like growth factor-I and bone: lessons from mice and men.
Pediatr. Nephrol. 24(7), 1277–1285 (2009)
47. Kawata, A., Mikuni-Takagaki, Y.: Mechanotransduction in stretched osteocytes–temporal
expression of immediate early and other genes. Biochem. Biophys. Res. Commun. 246(2),
404–408 (1998)
48. Klein-Nulend, J., Sterck, J.G., et al.: Donor age and mechanosensitivity of human bone
cells. Osteoporos. Int. 13(2), 137–146 (2002)
49. Kostenuik, P.J., Halloran, B.P., et al.: Skeletal unloading inhibits the in vitro proliferation and
differentiation of rat osteoprogenitor cells. Am. J. Physiol. 273(6 Pt 1), E1133–E1139 (1997)
50. Kousteni, S., Chen, J.R., et al.: Reversal of bone loss in mice by nongenotropic signaling of
sex steroids. Science 298(5594), 843–846 (2002)
51. Kousteni, S., Han, L., et al.: Kinase-mediated regulation of common transcription factors
accounts for the bone-protective effects of sex steroids. J. Clin. Invest. 111(11), 1651–1664
(2003)
52. Kveiborg, M., Flyvbjerg, A., et al.: Changes in the insulin-like growth factor-system
may contribute to in vitro age-related impaired osteoblast functions. Exp. Gerontol. 35(8),
1061–1074 (2000)
53. Leigh, M.W., Pittman, J.E., et al.: Clinical and genetic aspects of primary ciliary dyskinesia/
Kartagener syndrome. Genet. Med. 11(7), 473–487 (2009)
54. Leppanen, O.V., Sievanen, H., et al.: Pathogenesis of age-related osteoporosis: impaired
mechano-responsiveness of bone is not the culprit. PLoS One 3(7), e2540 (2008)
55. Li, J., Duncan, R.L., et al.: Parathyroid hormone enhances mechanically induced bone
formation, possibly involving L-type voltage-sensitive calcium channels. Endocrinology
144(4), 1226–1233 (2003)
56. Little, R.D., Carulli, J.P., et al.: A mutation in the LDL receptor-related protein 5 gene results in
the autosomal dominant high-bone-mass trait. Am. J. Hum. Genet. 70(1), 11–19 (2002)
57. Loots, G.G., Kneissel, M., et al.: Genomic deletion of a long-range bone enhancer
misregulates sclerostin in Van Buchem disease. Genome Res. 15(7), 928–935 (2005)
58. Marino, M., Galluzzo, P., et al.: Estrogen signaling multiple pathways to impact gene
transcription. Curr. Genomics 7(8), 497–508 (2006)
59. McAllister, T.N., Frangos, J.A.: Steady and transient fluid shear stress stimulate NO release
in osteoblasts through distinct biochemical pathways. J. Bone Miner. Res. 14(6), 930–936
(1999)
60. McDonald, F., Somasundaram, B., et al.: Calcium waves in fluid flow stimulated osteoblasts
are G protein mediated. Arch. Biochem. Biophys. 326(1), 31–38 (1996)
61. Miyauchi, A., Notoya, K., et al.: Parathyroid hormone-activated volume-sensitive calcium
influx pathways in mechanically loaded osteocytes. J. Biol. Chem. 275(5), 3335–3342
(2000)
62. Mullender, M.G., van der Meer, D.D., et al.: Osteocyte density changes in aging and
osteoporosis. Bone 18(2), 109–113 (1996)
63. Neidlinger-Wilke, C., Stalla, I., et al.: Human osteoblasts from younger normal and
osteoporotic donors show differences in proliferation and TGF beta-release in response to
cyclic strain. J. Biomech. 28(12), 1411–1418 (1995)
64. Nicolella, D.P., Feng, J.Q., et al.: Effects of nanomechanical bone tissue properties on bone
tissue strain: implications for osteocyte mechanotransduction. J. Musculoskelet. Neuronal
Interact. 8(4), 330–331 (2008)
65. Nishida, S., Endo, N., Yamagiwa, H., et al.: Number of osteoprogenitor cells in human bone
marrow markedly decreases after skeletal maturation. J. Bone Miner. Metab. 17(3), 171–
177 (1999)
66. O’Brien, C.A., Plotkin, L., et al.: Control of bone mass and remodeling by PTH receptor
signaling in osteocytes. PLoS ONE 3(8), e2942 (2008)
67. Ogita, M., Rached, M.T., et al.: Differentiation and proliferation of periosteal osteoblast
progenitors are differentially regulated by estrogens and intermittent parathyroid hormone
administration. Endocrinology 149(11), 5713–5723 (2008)
68. Orr, A.W., Helmke, B.P., et al.: Mechanisms of mechanotransduction. Dev. Cell 10(1),
11–20 (2006)
69. Ortner, D.J.: Aging effects on osteon remodeling. Calcif. Tissue Res. 18(1), 27–36 (1975)
70. Owan, I., Burr, D.B., et al.: Mechanotransduction in bone: osteoblasts are more responsive
to fluid forces than mechanical strain. Am. J. Physiol. 273(3 p2 1), C810–C815 (1997)
71. Parfitt, A.M.: The actions of parathyroid hormone on bone: relation to bone remodeling and
turnover, calcium homeostasis, and metabolic bone disease. Part III of IV parts; PTH and
osteoblasts, the relationship between bone turnover and bone loss, and the state of the bones
in primary hyperparathyroidism. Metabolism 25(9), 1033–1069 (1976)
72. Parfitt, A.M.: The cellular basis of bone turnover and bone loss: a rebuttal of the osteocytic
resorption–bone flow theory. Clin. Orthop. Relat. Res. (127), 236–247 (1977)
73. Passerini, A.G., Polacek, D.C., et al.: Coexisting proinflammatory and antioxidative
endothelial transcription profiles in a disturbed flow region of the adult porcine aorta. Proc.
Natl. Acad. Sci. USA 101(8), 2482–2487 (2004)
74. Pittenger, M.F., Mackay, A.M., et al.: Multilineage potential of adult human mesenchymal
stem cells. Science 284(5411), 143–147 (1999)
75. Plotkin, L.I., Aguirre, J.I., et al.: Bisphosphonates and estrogens inhibit osteocyte apoptosis
via distinct molecular mechanisms downstream of extracellular signal-regulated kinase
activation. J. Biol. Chem. 280(8), 7317–7325 (2005)
76. Quarto, R., Thomas, D., et al.: Bone progenitor cell deficits and the age-associated decline
in bone repair capacity. Calcif. Tissue Int. 56(2), 123–129 (1995)
77. Raab, D.M., Smith, E.L., et al.: Bone mechanical properties after exercise training in young
and old rats. J. Appl. Physiol. 68(1), 130–134 (1990)
78. Rath, A.L., Bonewald, L.F., et al.: Correlation of cell strain in single osteocytes with
intracellular calcium, but not intracellular nitric oxide, in response to fluid flow. J. Biomech.
43(8), 1560–1564 (2010)
79. Rawlinson, S.C., Pitsillides, A.A., et al.: Involvement of different ion channels in
osteoblasts’ and osteocytes’ early response to mechanical strain. Bone 19(6), 609–614
(1996)
80. Rickard, D., Harris, S.A., et al.: Estrogens and progestins. In: Bilezikian, J.P., Raisz, L.G.,
Rodan, G.A. (eds.) Principles of Bone Biology, vol. 1, pp. 655–676. Academic Press, San
Diego (2002)
81. Robling, A.G., Niziolek, P.J., et al.: Mechanical stimulation of bone in vivo reduces
osteocyte expression of Sost/sclerostin. J. Biol. Chem. 283(9), 5866–5875 (2008)
82. Robling, A.G., Turner, C.H.: Mechanical signaling for bone modeling and remodeling. Crit.
Rev. Eukaryot. Gene Expr. 19(4), 319–338 (2009)
83. Rosen, C.J.: Growth hormone, insulin-like growth factors, and the senescent skeleton:
Ponce de Leon’s Fountain revisited? J. Cell. Biochem. 56(3), 348–356 (1994)
84. Rubin, C.T., Bain, S.D., et al.: Suppression of the osteogenic response in the aging skeleton.
85. Rubin, J., Rubin, C., et al.: Molecular pathways mediating mechanical signaling in bone.
Gene 367, 1–16 (2006)
86. Rubin, J., Rubin, H., et al.: Constraints of experimental paradigms used to model the aging
skeleton. In: Rosen, C.J., Glowacki, J., Bilezikian, J.P. (eds.) The Aging Skeleton, pp. 27–
36. Academic Press, San Diego, (1999)
87. Ryder, K.D., Duncan, R.L.: Parathyroid hormone enhances fluid shear-induced [Ca2+i]
signaling in osteoblastic cells through activation of mechanosensitive and voltage- sensitive
Ca2+ channels. J. Bone Miner. Res. 16(2), 240–248 (2001)
88. Sakai, K., Mohtai, M., et al.: Fluid shear stress increases transforming growth factor beta 1
expression in human osteoblast-like cells: modulation by cation channel blockades. Calcif.
Tiss. Intl. 63(6), 515–520 (1998)
89. Sako, Y.: Effects of turbulent blood flow and hypertension on experimental atherosclerosis.
JAMA 179, 36–40 (1962)
90. Sawakami, K., Robling, A.G., et al.: The Wnt Co-receptor LRP5 is essential for skeletal
mechanotransduction but not for the anabolic bone response to parathyroid hormone
treatment. J. Biol. Chem. 281(33), 23698–23711 (2006)
91. Semenov, M., Tamai, K., et al.: SOST is a ligand for LRP5/LRP6 and a Wnt signaling
inhibitor. J. Biol. Chem. 280(29), 26770–26775 (2005)
92. Silbermann, M., Bar-Shira-Maymon, B., et al.: Long-term physical exercise retards
trabecular bone loss in lumbar vertebrae of aging female mice. Calcif. Tissue Int. 46(2),
80–93 (1990)
93. Speder, P., Noselli, S.: Left-right asymmetry: class I myosins show the direction. Curr.
Opin. Cell Biol. 19(1), 82–87 (2007)
94. Sterck, J.G., Klein-Nulend, J., et al.: Response of normal and osteoporotic human bone cells
to mechanical stress in vitro. Am. J. Physiol. 274(6 Pt 1), E1113–E1120 (1998)
95. Sutherland, M.J., Ware, S.M.: Disorders of left-right asymmetry: heterotaxy and situs
inversus. Am. J. Med. Genet. C Semin. Med. Genet. 151C(4), 307–317 (2009)
96. Tatsumi, S., Ishii, K., et al.: Targeted ablation of osteocytes induces osteoporosis with
defective mechanotransduction. Cell Metab. 5(6), 464–475 (2007)
97. Triplett, J.W., O’Riley, R., et al.: Mechanical loading by fluid shear stress enhances IGF-1
receptor signaling in osteoblasts in a PKCzeta-dependent manner. Mol. Cell Biomech. 4(1),
13–25 (2007)
98. Tsuji, T., Hughes, F.J., et al.: Effects of donor age on osteogenic cells of rat bone marrow in
vitro. Mech. Ageing Dev. 51(2), 121–132 (1990)
99. Turner, C.H., Forwood, M.R., et al.: Mechanical loading thresholds for lamellar and woven
bone formation. J. Bone Miner. Res. 9(1), 87–97 (1994)
100. Turner, C.H., Takano, Y., et al.: Aging changes mechanical loading thresholds for bone
formation in rats. J. Bone Miner. Res. 10(10), 1544–1549 (1995)
101. Turner, C.H., Warden, S.J., et al.: Mechanobiology of the skeleton. Sci. Signal. 2(68), pt3
(2009)
102. Umemura, Y., Ishiko, T., et al.: Effects of jump training on bone hypertrophy in young and
old rats. Int. J. Sports Med. 16(6), 364–367 (1995)
103. Vashishth, D., Gibson, G.J., et al.: Sexual dimorphism and age dependence of osteocyte
lacunar density for human vertebral cancellous bone. Anat. Rec. A Discov. Mol. Cell Evol.
Biol. 282(2), 157–162 (2005)
104. Villanueva, A.R., Sedlin, E.D., et al.: Variations in osteoblastic activity with age by the
osteoid seam index. Anat. Rec. 146, 209–213 (1963)
105. Wang, G.J., Sweet, D.E., et al.: Fat-cell changes as a mechanism of avascular necrosis of the
femoral head in cortisone-treated rabbits. J. Bone Joint Surg. Am. 59(6), 729–735 (1977)
106. Watson, C.S.: The Identities of Membrane Steroid Receptors: And Other Proteins Mediating
Nongenomic Steroid Action. Kluwer Academic Publishers, Boston (2003)
107. Weinbaum, S., Cowin, S.C., et al.: A model for the excitation of osteocytes by mechanical
loading-induced bone fluid shear stresses. J. Biomech. 27(3), 339–360 (1994)
108. Weinbaum, S., Guo, P., et al.: A new view of mechanotransduction and strain amplification
in cells with microvilli and cell processes. Biorheology 38(2–3), 119–142 (2001)
109. Yoshida, Y., Sue, W., et al.: The effects of augmented hemodynamic forces on the
progression and topography of atherosclerotic plaques. Ann. N. Y. Acad. Sci. 598, 256–273
(1990)
110. You, J., Jacobs, C.R., et al.: P2Y purinoceptors are responsible for oscillatory fluid flow-
induced intracellular calcium mobilization in osteoblastic cells. J. Biol. Chem. 277(50),
48724–48729 (2002)
111. You, J., Reilly, G.C., et al.: Osteopontin gene regulation by oscillatory fluid flow via
intracellular calcium mobilization and activation of mitogen-activated protein kinase in
MC3T3–E1 osteoblasts. J. Biol. Chem. 276(16), 13365–13371 (2001)
112. You, L., Cowin, S.C., et al.: A model for strain amplification in the actin cytoskeleton of
osteocytes due to fluid drag on pericellular matrix. J. Biomech. 34(11), 1375–1386 (2001)
The Effect of Aging on Skeletal
Mechanoresponsiveness: Animal Studies
Akhilesh A. Kotiya and Matthew J. Silva
Abstract Growth and remodeling of skeletal tissue in response to its mechanical

environment is a well established phenomenon. Relatively little is known
regarding the interaction of aging and skeletal responses to mechanical loading,
although several early studies have contributed to the ‘‘conventional wisdom’’ that
old bones are relatively unresponsive. Development of non-pharmacological
therapies for treatment of skeletal pathologies requires better understanding of
such interactions, especially if aimed at maintaining or restoring bone mass in the
elderly. The use of intrinsic (e.g., running) and extrinsic (e.g., tibial compression)
loading models provide means to study age effects in animal studies. We identified
15 studies that address age effects explicitly, although only nine of these include a
truly old group (e.g., 18–24 months old for mice). Though the outcomes of the
studies have not been uniform, two general themes have emerged. First, bones
from old animals are mechano-responsive provided they are presented with an
appropriate stimulus. Second, it is unclear if bones from old animals are less
responsive than from younger animals, as there is evidence for and against this
view. Therefore, we advocate a re-examination of the conventional wisdom, and
offer a few guidelines for designing studies to address the questions regarding
aging and bone mechano-responsiveness.
A. A. Kotiya (&)
Department of Biomedical Engineering, Dalhousie University,
Halifax, NS, Canada
e-mail: akhileshkotiya@gmail.com
M. J. Silva
Department of Orthopaedic Surgery, Washington University,
Saint Louis, MO, USA
e-mail: silvam@wustl.edu

DOI: 10.1007/8415_2012_115
Published Online: 1 May 2012
192 A. A. Kotiya and M. J. Silva
1 Introduction
The interdependence of skeletal form and function has been the subject of much
scientific inquiry. Early observations on bone growth and adaptation in response to
functional loading were made by Roux in 1881, and formally described by Wolff
in 1890. ‘Wolff’s law’ entails two important concepts regarding bone adaptation:
optimization of bone strength to weight, and remodeling of bone under the
influence of functional loading [1]. A proposal to put these observations regarding
bone adaptation into a quantitative framework was made by Carter [2], who noted
that cyclic strain history governs adaptation with different control mechanisms for
disuse versus overuse. Frost put forth a similar framework in the form of his
mechanostat hypothesis: bone adds mass if the habitual load increases above a
certain threshold and loses mass if the habitual load decreases below certain
threshold [3]. In the last few decades numerous experiments were carried out to
better understand how the different parameters of loading influence bone adapta-
tion. Some common ‘‘rules’’ have emerged from these experiments: the local strain
history is a key determinant of the tissue response; dynamic rather than static
strains drive bone adaptation; the dynamic strain magnitude required to initiate an
adaptive response decreases with increasing loading frequency; few loading cycles
are sufficient to trigger bone adaptation provided the strain magnitude is above an
adaptation threshold; bone reaches a new homeostasis state in response to altered
loading history and further loading at similar magnitude fails to invoke an addi-
tional response [4–6]. Although much progress has been made, most animal
studies have focused on external variables and have utilized young animals. We
have a very limited understanding of the influence of age, or other intrinsic
parameters, on loading-induced bone adaptation.
Skeletal physiology and/or bone mechano-responsiveness is potentially influ-
enced by a variety of systemic and local changes associated with aging. For
instance the number of osteocytes, the cell type proposed to be involved with bone
mechano-transduction, appears to decrease with aging [7–9]. The loss of muscle
mass and strength with age (sarcopenia) may also contribute to age-related bone
loss, either through common regulatory factors such as insulin-like growth factor 1
(IGF1) or simply because weaker muscles generate less skeletal loading [10, 11].
Aging is also associated with changes in vascular function and blood flow that
could potentially influence shear stress or chemotransport dependent transduction
mechanisms. Moreover, levels of various systemic hormones and local cytokines
are influenced by aging. Considering such systemic and local factors, it is natural
to inquire if and how aging affects the ability of bone to adapt in response to its
mechanical environment.
The issue of how aging affects skeletal responses to mechanical stimuli
carries clinical implications. After peak bone mass is attained in young
adulthood/maturity (*30 years age), net deficits in bone turnover are observed
in both men and women that result in progressive loss of bone mass. The loss
The Effect of Aging on Skeletal Mechanoresponsiveness 193
is expedited in menopausal women as a result of estrogen deficiency. This

often results in osteoporosis, a condition characterized by low bone mineral
density (BMD T-score below -2.5), deteriorated skeletal microstructure and
increased risk of fracture [12]. The condition affects approximately 44 million
people in United States, with an estimated 1.5 million osteoporotic fractures
per year and associated costs of $18 billion [13, 14]. Because the social and
economic cost associated with osteoporosis is high, there have been numerous
efforts to develop pharmacological therapies (see ‘‘Bisphosphonates and PTH
for Preventing Fractures’’ by Burr and Allen for a review of this topic). Briefly,
anti-catabolic therapies target osteoclastic resorption, while anabolic therapies
target osteoblast formation. The most commonly used class of drugs are bis-
phosphonates, which primarily act to block resorption. Although bisphospho-
nates have undisputed efficacy in reducing fracture incidence, recent concerns
have emerged about long-term ([5 years) treatment [15]. Currently, synthetic
parathyroid hormone 1-34 is the only FDA approved drug that has been shown
to result in anabolic bone modeling. But the duration of treatment with PTH is
limited to 2 years, as continued treatment was shown to increase chances of
osteosarcoma in rats [16]. Thus, there remains a need to develop additional
options to treat osteoporosis. Non-pharmacologic approaches that modulate
bone accrual and adaptation can potentially be harnessed to develop non-
invasive anabolic treatments with minimal side effects. Physical loading/
exercise is one approach. A better understanding of how aging influences
skeletal responses to loading is critical towards development of such a treat-
ment option(s).
The aim of this chapter is to provide an overview of animal studies that have
addressed the question of age-related changes in bone mechano-responsiveness.
By mechano-responsiveness we mean the net ability of bone to sense and respond
to any changes in its mechanical environment. If we consider a study comparing
two groups subjected to an identical loading environment (i.e., mechanical input),
group A is more mechano-responsive than group B if the magnitude of adaptation
(i.e., the response) in group A is greater than in group B. We do not address
possible underlying mechanisms that might relate to sensing, transduction and cell
function, each of which might be affected by aging and could contribute to changes
in bone mechano-responsiveness.
When examining the effects of aging, the ages at which comparisons are made
are of critical importance. As noted below, some studies have examined ‘‘aging’’
by comparing young versus mature animals, and few studies have conducted
studies using truly old animals. Aging is a continuum, but for simplicity we can
separate the lifespan into four distinct phases: young, mature, middle aged, old.
When describing the results from the literature, we have attempted to describe ages
in these terms, as summarized in Table 1. Our rationale for this is based largely
on the work of Harrison et al. at Jackson Labs [17].
Table 1 Ages that represent different phases in life for mice, rats and humans
Young Mature Middle Aged Old
Mousea 1–3 months 4–6 months 10–14 months 18–24 months
Ratb 1–3 months 5–7 months 12–16 months 21–28 months
Human 10–15 years 20-30 years 38–47 years 56–69 years
a
The mouse versus human comparison is based on Flurkey et al. [17] for C57Bl/6J mice, which
have a median lifespan of *26 months [61]. They state that ‘‘mice older than 24 months are
useful for pathology and life span studies; they are not as useful for normal aging studies’’. They
classify mature adult mice as 3–6 months, but we believe that 3 months is too young to be
considered ‘‘skeletally mature’’ and consider 5–6 months best for mature adult status [69].
Longevity data for other mouse strains is similar to C57Bl/6 but not necessarily identical [70]
b
The rat ages are based on simple linear interpolation from the mouse ages, assuming an average
longevity of *30 months in rats [71]. Others have recommended 9 months as an age at which
rats reach peak bone mass and is appropriate for interventions such as ovariectomy [72]
2 Overview of Approaches to Study Bone Responses

to Mechanical Loading Using In Vivo, Animal Models
A number of animal models have been developed to study the influence of mechanical
loading on bone. These models have been used to examine the effects of external
loading parameters such as strain magnitude, strain rate, frequency and number of
loading cycles. Generally these models involve alteration of the mechanical environ-
ment of the bone (usually, but not limited to, forelimbs and hindlimbs) and studying the
resulting changes in bone cell function/activity (e.g., bone formation rate), bone
structure and mechanical properties. Use of many of these models was reviewed by
Robling et al. [18], who emphasized that the objective of any loading model is to
generate or apply force that produces bone deformation, i.e., strain. Depending on the
mode of loading, these can be classified as intrinsic or extrinsic loading models (Fig. 1).
Intrinsic loading models utilize physiological activity that engenders contractile
muscle forces and joint reaction forces that result in bone loading. Typically, the
animal is trained to mimic certain forms of exercise (e.g., running, jumping,
swimming, climbing) that result in more strenuous loading (i.e., more loading
cycles or higher force) compared to the habitual loading environment of a cage-
dwelling lab animal. Being physiologic in nature the loading involves active
contributions of muscle and the associated effects (e.g., increased blood flow) on
bone and other tissues. However, there are limitations with intrinsic approaches:
the local loading (strain) parameters at the skeletal site of interest are difficult to
control in this type of model; the training regimens are not strictly ‘‘voluntary’’ as
the investigator places the animal in a setting that gives them strong incentive to
complete the activity and thus may produce physiological stress. On the other hand
extrinsic loading models allow one to exert better control over loading (strain)
parameters. Such extrinsic loading models make use of external devices to directly
load the skeletal segment of interest and thus alter the mechanical environment of
the bone. Most of our knowledge of the quantitative relationships between loading
and bone response is derived from studies using such extrinsic models [18].
Fig. 1 Models used in animal studies of skeletal responses to mechanical loading. (N non-
invasive, I invasive)
However, extrinsic loading models have their own limitations. For example, the
strain patterns, magnitudes and rates generated during such loading might not be
physiological. Also, active contribution of muscles and its associated physiological
changes are typically absent in these loading models. Both intrinsic and extrinsic
models can be further classified as invasive or non-invasive depending on the need
for surgical intervention. In the following section we give a brief overview of the
loading models commonly used to study responses to loading. We limit the review
to cover models that have been used in studies where age was a factor. Conse-
quently, swimming, climbing and vertebral compression are not discussed further.
Strain magnitude is the most widely accepted measure of local loading intensity of
relevance to in vivo studies of bone adaptation. Thus, it is important to describe loading
studies in terms of peak strain engendered by the loading protocol. For purposes of
comparison, peak principal strains on the tibia of humans have been reported in the range
of 500–1200 microstrain (le) for walking and running [19]. Strain values can be mea-
sured using electrical strain gages placed at a selected site(s) on the periosteum of the
bone of interest, typically in a small set of non-survival animals. Data are recorded during
the loading protocol and either reported as a function of activity for intrinsic studies, or a
function of externally applied force for extrinsic studies. This is the classic approach and
remains the gold standard [19]. Nonetheless, the main limitation of this approach is that
the gage is not necessarily placed at the site of maximal (or minimal) strain on the bone,
and represents a single descriptor of a complex strain state. Alternate image-based
approaches have been described which capture some of the spatial variation in bone
surface strains (e.g., on the mouse tibia during axial compression [20]). In addition, finite
element models have been used to supplement the experimental strain data, and these can
add greatly to the understanding of the strain state in the bone of interest [21–24].
3 Intrinsic Loading Studies
3.1 Running
Running is a commonly used method of intrinsic skeletal loading. Some studies

utilize voluntary wheel running, although most use treadmill running because
it allows greater control over the loading regimen. (For rodents on treadmills,
ambulation is probably a more accurate term than ‘‘running’’, as there is at least
one paw in contact with the ground at speeds commonly used.) The loading
magnitude can be increased by either attaching additional weight to the animals or
increasing the inclination or speed of the treadmill. The rate of loading and number
of loading cycles are controlled by the speed of the treadmill and the duration of
training. There are few data on bone strains engendered in rats by running, but the
available data indicate similar values as in humans. In growing male rats, ulnar
strains of 700–1200 le were recorded by Mosley et al. [25] during running on a
level surface. In addition, tibial strains of approximately 700 le were reported by
Rabkin et al. [26] during treadmill running, with a modest increase (+13%) in
strain magnitude with a doubling of speed (from 8 to 16 m/min). The influence of
age on strains produced during running is unclear. Indrekvam et al. [27] reported
that peak strain on the femoral diaphysis during treadmill running did not differ
significantly between rats aged 1.5, 3 and 12 months. Similarly, Keller and
Spengler [28] reported no significant changes in strain magnitude on the femur of
running rats from 1.5 to 7 months age, despite a threefold increase in body weight
during this interval. Thus, available data indicate no change in bone strain
engendered by running during growth and maturation, although we could find no
data on changes from maturity to old age.
Raab et al. used treadmill running to examine the influence of aging on
mechano-responsiveness [29] (Table 2). They evaluated the effect of 10 weeks of
treadmill running on fat-free weight and mechanical properties (as determined by
three-point bending) of femur and humerus of young (2.5 months) and old
(25 months) female rats. Compared to sedentary control, running increased the fat-
free weight of the femur in both young and old rats by a similar amount. Trained
rats of both age groups also had significantly greater ultimate force in the femur
and humerus and greater yield force in the humerus. However the moment of
inertia of the humerus and femur (mid-diaphysis) was unchanged. It was con-
cluded that the bone adaptation in response to treadmill running was similar in
young and old rats, even though the old rats ran at a slower speed (young: 36 m/min;
old: 15 m/min) and thus may have had a lower magnitude strain stimulus.
Leppanen et al. made similar observations on comparing the response of mature
(11 months) and old (22 months) female rats to treadmill running [30]. Compared
to the sedentary control, 14 weeks of treadmill running increased the ultimate force
of the femur (mid-diaphysis) of old rats but not mature rats. Changes at the femoral
neck, a common site of osteoporosis related fractures, were also studied. In older
rats running increased the ultimate force (determined by compression test), the
cross-sectional area and bone mineral content (BMC) at the femoral neck. Running
did not influence these parameters in mature rats. The values of these parameters
were similar for trained old rats and sedentary mature rats suggesting that running
was not anabolic but prevented bone loss in older animals. Also exercise had no
influence on the tibial metaphysis in either age group, implying that the bone
adaptation is site specific in this model. In the same study, Leppanen et al. reported
on treadmill running effects in mature (11 months) and old (19 months) male rats.
Table 2 Summary of studies that examined the influence of age on bone responses to intrinsic loading
Study Animal sex, Age Protocol Significant observations regarding Conclusion
species (strain) the influence of loading
Raab et al. Female, rat 2.5 months— Treadmill running, 10 weeks, Fat free weight: Femur—Y:, O: No age effect
[29] (Fisher 344) young 60 min/day, 5 days/week, Ultimate force: Femur—Y:, O:,
25 months—old 15 deg incl. Humerus—Y:, O:
Young—36 m/min, Old— Yield force: Humerus—Y:, O:
15 m/min
Leppanen Female, rat 11 months—mature (FM) Treadmill running, 14 weeks Ultimate Force: Femur—FO:, No age effect
et al. (Sprague- 22 months—old (FO) 30 min/day, 30° inclination Femoral neck—FO:, MO:
[30] Dawley) Cross-sectional area: Femur—
Male, Rat 11 months—mature (MM) MM:, Femoral neck—FO:,
(Sprague- 19 months—old (MO) MO:
Dawley) BMC: Femoral neck—FO:, MO:
Bennell Female, Rat 1 months—young Treadmill running, 12 weeks Ultimate force: Tibia—Y:, IM: No age effect
et al. (Sprague- 4 months—immature 60–70 min/day, 5 days/ Cross-sectional area: Tibia—Y:,
[31] Dawley) week, 26 m/min IM:
Histomorphometric indices:
Tibia—Y:, IM:
The Effect of Aging on Skeletal Mechanoresponsiveness
BMC: Lumbar (L1–L4)—Y:, IM:

Umemura Female, Rat 3 months, 6 months— Treadmill running, 8 weeks Fat free weight: Tibia—Y,M :; Age effect; old not
et al. (Fisher 344) young 60 min/day, 5 days/week, Femur—Y: responsive
[32] 12 months—mature 30 m/min Diameter: Femur—Y:
20 months, Jump training, 8 weeks Fat free weight: Tibia—All:; Modest age effect;
27 months—old 100 times/day, 5 days/week, Femur—Y:, O: old less
40 cm (up) Diameter: Femur—Y:, responsive
Tibia—All:
Jarvinen Male, Rat 1 months—young Treadmill running, 14 weeks Femur [1] No age effect
et al. (Sprague- 8 months—mature 10 min/day, 4 days/week, Ultimate force: Y:, M:
[33] Dawley) 18 m/min, 30° inclination Cross-sectional area: Y:
BMC: Y:, M:
197
(continued)
Table 2 (continued)
198

species (strain) the influence of loading
Hoshi et al. Female, 2–18 months Wheel running, at ages: Femur Modest age effect;
[34] Mice (ICR) 10–30 week—Young Ultimate force: YE:, ME:, LE: old less
Exercise (YE) Cortical thickness: YE:, ME:, responsive
30–50 week—Mature LE:
Exercise (ME) Bone density: All:
50–70 week—Old
Exercise (OE)
10–70 week—Life long
Exercise (LE)
Honda Male, Rat 3 months—young Jump training, 8 weeks BMC: All: No age effect
et al. (Wistar) 11 months—mature 10 times/day, 5 days/week, Cortical area, moment
[36] 40 cm of inertia: All:
Buhl et al. Male, Rat 4 months—young Squat-like training, 9 weeks Tibia Modest age effect;
[35] (Fisher 344) 12 months—mature 50 times/day, 3 days/week Trabecular spacing, BV/TV: O: old more
22 months—old added weight—65% body responsive
weight
: Increases in response to loading, Y Young, IM Immature, M Mature, O Old. [1] values were adjusted for muscle mass
A. A. Kotiya and M. J. Silva
Although most of the adaptive responses were similar in males and females, some
sex differences were observed. An increase in mid-femoral cross sectional area
in mature male rats and loss of body weight in both mature and old male rats
was observed in response to exercise; female rats did not show these changes.
On the other hand, exercise increased the femoral length of mature female but not
male rats. The differences in response indicate that the sex of the animals must be
considered when studying bone adaptation under mechanical loading.
Bennell et al. [31] studied the influence of 12 weeks of treadmill running on
skeletal adaptation in female rats at two ages: 1-month old (rapid growth phase)
and 4-month old (steady growth phase). Compared to sedentary controls, running
increased bone formation indices, bone size and mechanical properties in the tibias
of both age groups. Similar findings were noted in the lumbar spine (L1–L4).
On the other hand, the tibial metaphysis and femur did not show any changes in
response to exercise. In summary, there were no age-related differences in the
effects of running on bone properties at the local level. The only difference in
response to running between the two age groups was that trained 1-month old rats
had greater gain in total body bone area and BMC than trained 4-month old rats.
Nonetheless, because both of these ages are still early in the rat lifespan, this study
addresses loading during growth rather than during aging.
Contrary to the aforementioned findings of no age-related decline in respon-
siveness to running, Umemura et al. reported that older rats are less responsive to
treadmill running than younger rats [32]. In young animals (3 and 6 months),
8 weeks of running increased the fat-free weight of the femur and tibia, length and
diameter of the femur, and length of the tibia. At a mature age (12 months), the
only significant effect of run-training was an increase in tibial weight. At older
ages (20 and 27 months), there were no differences in these parameters between
run-trained and control animals, indicating that the older animals did not respond
to treadmill running. The authors noted that because they imposed the same
running speed (30 m/min) on all age groups, ‘‘it is considered that the intensity
was too high for the old rats’’. We note that the outcomes in this study were limited
to relatively simple measures of bone geometry and mass; it is possible that other
measures (e.g., local bone structure, bone mechanical properties) may have been
more sensitive to running.
Jarvinen et al. observed differences in adaptive mechanism between young and
mature rats in response to treadmill running [33]. After 14 weeks of treadmill
running, both young (1 month) and mature (7 months) animals had significantly
higher BMC, BMD and breaking load at the femoral neck. However, only the
young animals had a significant increase in femoral neck cross-sectional area
(+25%) compared to a non-significant change (+10%) in mature rats. By contrast,
young rats had a smaller increase (+11%) in BMD than mature rats (+23%). Run
training did not have any influence on the length of the femur for either age group.
It was concluded that growing animals respond primarily by changes in bone
size (increased area) whereas mature animals respond primarily by changes in
bone density. We note that this conclusion is not strongly supported by the data,
as the increases with running did not differ significantly between age groups.
These authors also looked at how aging influences the skeleton’s ability to
maintain treadmill running induced gains during de-conditioning (free cage
activity only). Both the age groups lost bone after 14 weeks of de-conditioning. It
was concluded that there is a need of constant exercise for maintenance of bone
gained in response to exercise during growth and maturation.
Hoshi et al. [34] studied the effect of running exercise during different stages of
life on femoral density and mechanical properties. Ten-week old mice were
subjected to voluntary running in a revolving wheel either at different ages
(10–30, 30–50 and 50–70 weeks) or throughout the duration of the experiment
(10–70 week). It was observed that voluntary running distance decreased with age.
However, compared to sedentary control, running at all ages—with the exception
of the 50–70 week old animals—increased the cortical thickness index, breaking
force, and ultimate stress and elasticity of femurs. Bone density was higher in all
the mice that were subjected to exercise. Based on these observations it can be
concluded that exercise at any age from youth to middle age is beneficial in
arresting age-related bone loss, although starting exercise at middle age may be
less effective. The latter conclusion may be influenced by the reduction in distance
run in the older animals.
3.2 Jumping
Jump training is another physiologically relevant, intrinsic loading modality. The

method involves training the animals to jump up to or down from a platform.
Mosley et al. [25] reported a strain magnitude of 2300 le for dropping from a
30 cm height (*twofold greater than for running). The strain can be further
increased by attaching additional weight to the animal while the number of loading
cycles is determined by the number of jumps. The rate of loading cannot be easily
controlled for this loading method. We are unaware of studies that have charac-
terized age-related differences in strain magnitude or distribution at relevant
skeletal sites for jumping. A variation of this loading modality is squat-like
exercise [35], although it is not known if the resulting strain magnitude is similar
to that of jumping.
Umemura et al. subjected female rats of different ages (3–27 months) to
8 weeks of jump training [32] (same study reviewed for run training in previous
session). The fat-free weight of the tibia and femur of jump-trained rats was
greater than untrained rats for all age groups (except for the 12-months femur), and
the diameter of the tibia (measured at the distal tibio-fibular junction) was also
greater in jump-trained rats for all age groups. However, jump training increased
femoral and tibial length and tibial diameter only in the younger groups (3 and
6 months). Compared with the observations in rats subjected to running, it was
concluded that jump training has a greater effect on bone adaptation than running
despite the briefer training time (jumping 10 min/day; running 60 min/day). It was
speculated that this difference is a result of jump training’s ability to induce higher
peak strains, consistent with Frost’s mechanostat hypothesis. Importantly, the

effects of jump training on bone hypertrophy were not limited by age.
Honda et al. also observed that jump training has skeletal benefits that accrue
(and are preserved) independent of age [36]. In both young (3 month) and mature
(10 month) rats, 8 weeks of jump training increased tibial BMC and BMD
[determined by dual-energy X-ray absorptiometry (DXA) every 4 weeks] and
strength (determined by three-point bending) compared to sedentary controls. This
bone accrual was maintained throughout the de-conditioning duration of 24 weeks,
again irrespective of age. It was concluded that aging does not influence the ability
of skeleton to respond to exercise or the maintenance of skeletal benefits induced
by jump training. Tibial parameters measured ex vivo (BMC, ultimate force,
cortical area, medullary area, endosteal perimeter, periosteal perimeter, fat-free
weight, maximum and minimum moment of inertia) were greater in the exercise
group at the end of the de-conditioning period compared to the sedentary group.
Exercise effects were not different between age groups, except for periosteal
perimeter which was increased more in the mature rats than in the young. Thus,
there was no loss of responsiveness with aging.
Buhl et al. [35] studied the influence of squat-like training (a similar loading
modality as jumping) on young (4 months), mature (12 months) and old (22 months)
male rats. While the above studies found that older animals are as responsive to jump
training as younger animals, Buhl et al. concluded that aged bones actually display
higher responsiveness to loading. Training had no influence on mechanical proper-
ties of the femur (determined by three-point bending) or on femoral cross-sectional
area, cortical area and moment of inertia in any age group. In addition, loading did not
influence mineral apposition rate and bone formation rate at the tibial diaphysis.
However, the tibial metaphysis of older animals subjected to loading had signifi-
cantly lower trabecular spacing and greater BV/TV and trabecular number
(non-significant) than the control animals. Training also increased the medullary area
of the femur in older animals. Based on these observations it was concluded that the
training, though it failed to benefit young and mature bones, was marginally bene-
ficial to old bone.
3.3 Summary: Bone Responses to Intrinsic

Mechanical Loading
We found only eight studies of intrinsic mechanical loading that directly compared
different ages, and only five of these include a true old age group (Table 2). Of
these eight studies, the majority found no effect of age. Therefore, they support the
view that the ability of the skeleton to adapt to altered mechanical loading is not
compromised by aging. The only study that found a clear, negative influence of
age was the running protocol of Umemura et al. [32], wherein young rats had
favorable bone responses to treadmill running while older rats did not respond.
Interestingly, in the same study both young and old rats responded favorably to
jump training with only modest evidence of an age effect. Because the jump
training likely produced higher magnitudes of bone strain (and strain rate), their
results suggest that young bones are more responsive than old bones to low-strain
loading but if the strain exceeds an ‘‘adaptation threshold’’ both age groups are
responsive by a similar amount. On the other hand, Leppanen et al. [30] conducted
a very thorough evaluation of the effects of treadmill running in old rats and
reported that there were favorable bone responses. One notable difference between
these studies is the treadmill inclination angle (none reported for Umemura et al.,
versus 30° for Leppanen et al.). It is possible that making the animals run uphill
increases the bone strain sufficiently to exceed the putative threshold.
4 Extrinsic Loading Models
4.1 Functionally Isolated Avian Ulna
Rubin and Lanyon [37, 38] greatly advanced the science of bone adaptation using
an animal model that allows for complete control of loading history during the
experimental period. In this model an 11 cm segment of the ulnar diaphysis of
turkeys is functionally isolated by removing the distal and proximal ends and
placing caps at each end, held in place with transverse pins through which loads
are applied. An external fixator prevents incidental loading of the bone segment.
Controlled static loading is applied by springs, while controlled dynamic loading is
applied using a materials testing machine. A strength of the model is that the
loading history is completely controllable while maintaining the bone in situ with
muscle, nerve and vascular attachments intact. Limitations include the drastic
change in normal loading environment (which is likely to be ‘‘perceived’’ by the
bone as a disuse state), and the invasive surgical procedure, which may produce
local or systemic responses that influence bone adaptation (e.g., a ‘‘regional
acceleratory phenomenon’’). Such non-voluntary, invasive loading models have
also been developed to study effect of loading on caudal (tail) vertebrae in rats [39]
and lumbar vertebrae in rabbits [40].
Related to aging, Rubin et al. observed an age-related decline in the response to
mechanical stimulus in male turkeys using the functionally isolated ulnar loading
model [41]. Compared to non-loaded control, 8 weeks of cyclic loading (3000 le,
300 cycles/day) increased the cortical area of the loaded bone in young adult
(mature; 1 year) but not old (3 year) animals (Table 3). The greater cortical area in
the mature group was a result of greater periosteal area and lower endosteal area.
Also, compared to non-loaded controls the mineral apposition rate was greater in
the loaded bone of mature animals but not old animals. Based on these observa-
tions it was concluded that a loading stimulus that is ‘‘clearly osteogenic in the
young adult skeleton is hardly acknowledged in older bone tissue’’, indicating a
decline with age in the ability of bones to sense and respond to loading.
Table 3 Summary of studies that examined the influence of age on bone responses to extrinsic cyclic loading
species (Strain) the influence of loading
Rubin et al. Male, Turkey 12 months—young Ulna isolation, axial loading, 8 weeks, Cortical area: Y : Age effect;
[37] 36 months—old 300 cycles/day, 3000 le, 2 Hz Mineral apposition rate: Y : old not
responsive
Turner et al. Female, rat 9 months—mature Tibial 4-point bending, 2 weeks, 36 Periosteal woven bone formation Age effect;
[44, 45] (Sprague- 19 months—old cycles/day, 2 Hz load [40 N: M: (100%), old less
Dawley) Mature: 27–64 N (1400–3000 le) O: (60%) responsive
Old: 30–64 N (1600–3100 le) Endocortical lamellar bone
formation
load 40 N: M:; load 64 N: M::,
O:
Srinivasan Female, mice 4 months—younga Tibial cantilever bending, 2 weeks, Bone formation rate: Y::, O: Age effect;
et al. (C57Bl/6) 21 months—old 50–250 cycles/day, 1200–2400 le, (O is 2.5-fold less than Y) old less
[53] 1 s load±10 s rest O: rest insertion increases bone responsive
formation similar to doubling of
strain
Kesavan Female, mice 2 months, 4 Tibial 4-point bending, 2 weeks, Cross sectional area: Y:, M: No age effect
The Effect of Aging on Skeletal Mechanoresponsiveness
et al. (C57Bl/6, months—young 36 cycles/day, 9 N (*3800 le), 2 Hz BMC: Y:, M:

[46] C3H/He) 8 months—
mature
Silva and Male, mice 4 months Tibial 3-point bending, 2 weeks, Endocortical bone formation: No effect of
Brodt (SAM) SAMP6 (senescent) 60 cycles/day, 1000–2000 le SAMR1:, SAMP6: ‘‘senescence’’
[51] SAMR1 (control) endocortical,
0.5 s load ? 10 s rest
(continued)
203
Table 3 (continued)
204

species (Strain) the influence of loading
Brodt and Male, mice 7 months—mature Tibial axial compression, 1 week, Endocortical bone formation: M :, Modest age effect;
Silva (BALB/c) 22 months—old 60 cycles/day, 900–1900 le O :: old more
[57] endocortical, 1400–3100 le Periosteal bone formation: M :, O : responsive
periosteal, Dose–response for both M, O (endocortical)
0.5 s load ? 10 s rest
Silva et al. Female, mice 2, 4 months— Tibial axial compression, 6 weeks, Cortical bone volume: Y :, M :, MA : Modest age effect;
[58] (BALB/c) young 60 cycles/day, 1300 le endocortical, Trabecular bone volume: M;, MA; young more
7 months—mature 2400 le periosteal, 0.5 s load ? 10 s responsive
12 months— rest
middle age
Lynch et al. Female, mice 2 months—young Tibial axial compression, 2 weeks, 1200 Cortical bone area, mom. inertia: Age effect;
[59, 60] (C57Bl/6) 6 months—mature cycles/day, 1200–2200 le, 4 Hz Y :, M : young more
Trabecular bone volume: Y ::, M : responsive
: Increases in response to loading, :: Relatively larger increases in response to loading, ; Decreases in response to loading; Y Young, M Mature, MA Middle
Age, O Old. Strains are periosteal strains unless noted otherwise
a
Described in discussion only
A. A. Kotiya and M. J. Silva
4.2 Four-Point Tibial Bending
Four-point bending of the rodent tibia, as described for the rat by Turner et al. [42]
and for the mouse by Akhter et al. [43], was perhaps the first non-invasive,
extrinsic loading approach used to study bone adaptation. With the animal
anesthetized, the hindlimb (tibia) is supported on its medial surface by two sup-
ports a certain distance apart. An extrinsic load transverse to the long axis of the
bone is applied at two points (between the supports) on the lateral surface. The
hindlimb is held in position by securing the foot. Bone between the two load points
is analyzed to study adaptation in response to loading. If the loading fixtures are
connected to a materials testing system, then the load magnitude, rate and number
of loading cycles can be precisely controlled. As with other extrinsic loading
models, a force-strain calibration should be done a priori to determine the local
strain magnitude at the site of interest. Among the drawbacks of this model are that
it loads only cortical bone, and that transverse loading of the tibia can result in
periosteal bone formation related to contact pressure rather than bone bending.
A ‘‘sham bending’’ load case (where the loading and support points are placed
directly opposite each other) can be used to correct for this effect, and endocortical
results appear to be unaffected by periosteal contact. Nonetheless, most investi-
gators no longer use this model because of the confounding effects of local contact
near the site of interest.
Turner et al. [44, 45] used this model in separate studies to examine the
influence of age on bone adaptation and concluded that aging increases the loading
threshold needed to trigger bone formation. Endocortical lamellar bone formation
in mature rats (9 months) showed a dose response to loading for loads above 40 N
(peak periosteal strain *2000 le, 36 cycles/day, 2 weeks). In contrast, a much
higher load (64 N; *3100 le) was required to induce an increase in endocortical
lamellar bone formation in middle-aged/old (19 month) rats. Moreover, the bone
formation rate at the 64 N load was fivefold less for 19-month old rats compared
to 9-month rats. The lower indices were observed in spite of almost equal
endocortical strain engendered by loading in mature and middle-aged animals
(i.e., similar force-strain calibrations). Loading resulted in periosteal woven bone
formation in 100% of mature (9 month) rats but only 60% of middle-aged/old
(19 month) rats at load magnitudes greater than or equal to 40 N. Based on the
endocortical and periosteal results, it was concluded that increasing age reduces
the mechano-responsiveness of bone.
On the contrary, Kesavan et al. studied bone adaptation in response to four-point
bending in two different strains of mice (C57Bl/6, C3H/He) at different ages
(2, 4 and 8 months) and concluded that age does not influence bone response to
loading [46]. Compared to the non-loaded limb, loading increased total area, total
mineral content, endosteal perimeter and periosteal perimeter (determined by
diaphyseal pQCT) of the loaded limb in both young (2, 4 month) and mature
(8 month) mice of each strain, indicating that age did not influence bone adaptation
in this model. However, middle-aged/old mice were not included in this study.
An alternative approach to the use of chronologically aged animals is the use of

mice that have an early aging (senescence) phenotype. Several such models relevant
to bone have been described. We utilized the P6 strain of the senescence accelerated
mouse (SAMP6), which has features reminiscent of age-related osteopenia. For
example, as early as 4 months age SAMP6 mice have enlarged periosteal and en-
docortical diameters, reduced trabecular bone volume, reduced endosteal bone
formation rate, reduced marrow osteogenesis, and reduced resistance to fracture
(fracture energy) compared to SAMR1 control mice [47–50]. Nevertheless, when
loaded by in vivo tibial three-point bending (1000–2000 le, 60 cycles/day, 2 weeks)
endocortical bone formation was activated by a similar amount in both SAMP6
(senescent) and SAMR1 (control) mice. We concluded that there was ‘‘little evi-
dence of diminished responsiveness to loading in the SAMP6 skeleton’’ [51]. The
use of the SAMP6 mouse is not a substitute for studies of chronologically aged
animals, and we no longer advocate use of this model [52]. Nonetheless, the results of
our loading study support the view that responses to loading are not necessarily
compromised by a relatively low level of baseline bone formation or pre-existing
osteopenia, conditions that occur with aging.
4.3 Tibial Cantilever Bending
Gross et. al developed a cantilever bending model that applies a transverse load to
the distal end of the tibia while the proximal end at the knee joint is clamped [22].
The strain at the mid-diaphysis is modulated by controlling the magnitude of the
applied load. A strength of the model is that it does not involve direct contact
between the loading surface and the bone surface of interest (in contrast to four-
point tibial bending). One limitation is that the applied load is not transmitted
through the metaphyseal region and hence the modality does not lend itself to
study trabecular bone adaptation. The loading direction is different from that
associated with normal physiological activities, which might be viewed as a
strength or limitation.
Srinivasan et al. [53] applied cantilever tibial bending to old (22 month) mice
and concluded that it was possible to induce an anabolic bone response, but that old
mice were less responsive than young (4 month) mice. (The results of the young
mice are mentioned in the Discussion only, and appear to have been obtained in a
separate, previous experiment by the same authors.) The rate of periosteal bone
formation induced in old mice by ‘‘low-magnitude’’ rest-inserted loading (1200 le,
50 cycles/day, 10 s rest interval between load cycles) was nearly 2.5-fold less than
that induced by a similar protocol in young mice. In addition, old mice did not
demonstrate a further increase in bone formation rate when loading magnitude was
doubled (2400 le), contrary to findings in young animals. The authors suggested
that a deficit in the number of available osteoblasts that can be activated might be
the reason for the inability of old mice to respond to higher loads.
4.4 Axial Tibial Compression
Axial compression of the hindlimb was developed separately by Fritton et al. [54]
and de Souza et al. [55] as a non-invasive method to stimulate bone formation in
the mouse tibia. The hindlimb of the animal is secured between supports at the
knee and foot; controlled compressive load is applied through these supports, in
the direction of the long axis of the tibia. Since the supports do not contact the
periosteal surface of the bone, adaptation of the entire length of the tibia including
cortical bone at the diaphysis and trabecular bone at the proximal metaphysis can
be studied. Strain gage and finite element analysis methods have been used to
characterize force-strain relationships and strain distributions at the mid-diaphysis
[55, 56]. Owing to the natural curvature of the tibia, a combined compression-
bending loading state is generated in the mid-diaphysis. This results in compres-
sive strains near the postero-lateral apex of the tibial cross-section with tensile
strains on the antero-medial flat. However, characterization of strain distribution
in trabecular region of metaphysis under applied loading remains a challenge.
Currently, this is one on the most popular extrinsic models used to study bone
adaptation.
We compared the response of mature (7 months) and old (22 months) male,
BALB/c mice to axial tibial compression, with a focus on cortical bone [57].
BALB/c mice represent an intermediate bone mass strain. Legs were loaded at one
of three force levels (range 900–1900 le endocortical, 1400–3100 le periosteal;
60 rest-inserted cycles/day, 5 days). Mice from both age groups showed a strong
anabolic response at the mid-diaphysis. At the endocortical surface, aged mice had
a significantly greater response to loading than mature mice while responses at
the periosteal surface did not differ between age groups (Fig. 2). We concluded
that aging does not limit the short-term anabolic response of cortical bone to
mechanical stimulation in this animal model.
In a follow-up study, we examined female, BALB/c mice ranging in age from
young to middle-aged (2, 4, 7, 12 months) [58]. Using analysis of serum and bone
mRNA in mice not subjected to loading, we noted an age-related decline in
markers of bone formation, corresponding with the transition from growth to
skeletal maturity. We then performed axial tibial compression (*1300 le
endocortical, *2400 le periosteal; 60 rest-inserted cycles/day, 3 days/week) and
evaluated changes in gene expression by qRT-PCR after 1 week of loading. Bone
formation related genes [e.g., type 1 collagen (Col1a1), osteocalcin (Bglap)] were
upregulated in an age-dependent manner; younger mice did not show evidence of
an increase whereas the expression in the loaded tibias of older mice increased to
levels seen in young mice. Finally, we performed 6 weeks of loading in another set
of young to middle-aged mice and followed changes in bone structure by in vivo
microCT. Loaded tibias in each age group had significantly greater cortical bone
volume (BV) than contralateral control tibias, due to relative periosteal expansion.
The loading-induced increase in BV was greatest in 4-month old mice, suggesting
Fig. 2 a Fluorescent photomicrographs of mid-diaphyseal tibial sections from loaded and

control tibias of 22- and 7-months mice. Samples were collected on day 11 following tibial
compression on days 1–5, and fluorochrome labeling on days 5 (green) and 10 (red). An increase
in endocortical and periosteal labeled surface is evident in loaded tibias of both ages compared to
controls. b Bone formation rate measured from tibial sections (n = 8–13 mice/group) from
the same study. There is no loss of mechanoresponsiveness in the old mice. (From Brodt and
Silva [57])
that this age was most responsive to loading. Unexpectedly, trabecular bone
volume fraction (BV/TV) was reduced in loaded limbs compared to controls in
4, 7 and 12 month groups, and was unchanged with loading in the 2 month group.
We concluded that, at cortical sites mechanical loading can overcome the normal,
age-related decline in bone formation in mice, with some evidence that the
young-adult skeleton (4 months) is more responsive than the mature to middle-
aged skeleton (7–12 months).
Lynch et al. recently used the axial tibial loading model in two separate studies
to compare bone adaptive responses in young, growing (2 months) versus mature,
adult (6 months) mice [59, 60]. Female C57Bl/6 mice were subjected to 2 weeks
of daily loading (1200 or 2200 le periosteal; 1200 cycles/day, 5 days/week)
and morphology of cortical and trabecular bone was assessed by post hoc
microCT. Comparisons between ages were made challenging because of different
relationships between applied force and peak periosteal strain measured at the
mid-diaphysis, leading the authors to perform strain-matched and load-matched
comparisons. Old mice were not responsive when loaded at a force (5.9 N) that
produced a strain magnitude of 1200 le periosteal, while young mice were
responsive when loaded at a force (11.5 N) that produced the same strain. By
contrast, when old mice were loaded at 11.5 N (2200 le) they had a positive
anabolic response that was similar in magnitude to the young mice at the diaphysis
(cortical) but less than the young mice at the metaphysis (trabecular). Notably,
trabecular BV/TV was increased in both age groups, contrary to our findings of
trabecular bone loss with loading [57, 58]. (Ongoing work in our lab suggests that
the difference in trabecular responses is attributed to the waveform/cycle number
differences between our protocol versus the protocol used by Lynch et al.) These
authors concluded that loading was anabolic for cortical and trabecular bone in
adult mice, albeit to a lesser degree than in young, growing mice.
4.5 High-Frequency Low-Magnitude Vibration
Loading modalities using high-frequency, low-magnitude vibration have gained

increasing attention [61]. It is believed that this kind of loading mimics the muscle
forces induced on bone during postural activities such as standing. Huang et al.
[62] reported that the 30–50 Hz component of postural muscle activity is altered
with aging, perhaps contributing to age-related bone loss. It was hypothesized that
extrinsic vibrational loading might compensate for this decline, and either stim-
ulate bone formation or prevent bone loss. One common method to apply this type
of loading is to place the animal (or human subject) on a platform that oscillates
vertically and subjects the animal to whole-body vibration (WBV). The loading
magnitude and frequency of the platform can be precisely controlled. However if
the animals (esp. rodents) are allowed to move freely over the platform, consistent
strains at a particular skeletal site cannot be guaranteed.
WBV loading has been reported to be mildly anabolic in young rodents [63, 64],
but it has not been widely studied in aged animals. We examined the influence of
WBV on bone in mature (7 months) and aged (22 months) male, BALB/c mice. As
reviewed above, these mice exhibit anabolic cortical responses to low-frequency,
high-magnitude tibial compression [57]. In contrast, we observed that WBV had
minimal influence on cortical bone at tibial mid-diaphysis or trabecular bone at the
proximal tibia for both mature and old mice [65]. Loading increased lower leg BMC
(determined using DXA) in adult mice but not in old mice. However, since most
outcomes (microCT, dynamic histomorphometry) did not show any loading
induced changes, we concluded that both the mature and old mice are unresponsive
to WBV.
4.6 Summary: Bone Responses to Extrinsic

Mechanical Loading
We found only seven animal studies of extrinsic mechanical loading that directly
compared different ages, and only four of these included a true old age group
(Table 3). Of these four, three reported a negative influence of age and one a
neutral-to-positive influence. We note that three of the four studies did report an
anabolic response in old animals, indicating that it was possible to elicit a
mechanoresponse, although perhaps diminished. Taken together, the limited
available data indicate that with aging there is a decline—but not a loss—in
skeletal mechanoresponsiveness to extrinsic loading.
5 Skeletal Unloading
Whereas increased skeletal loading can stimulate bone formation and increase
bone mass, diminished skeletal loading can stimulate bone resorption and lead to
decreased bone mass, sometimes called disuse osteopenia. Astronauts, paraplegic
and quadriplegic patients, and trained athletes after de-training all demonstrate
significant bone loss as a result of reduction in the skeleton’s functional demands.
Unloading related bone loss is more pronounced in skeletal sites that experience
higher habitual loads and/or sites that are closer to the ground, perhaps because of
fluid pressure effects [6]. Also trabecular bone shows much more rapid loss
compared to cortical bone. A better understanding of unloading related bone loss
would help us to develop therapies to minimize such bone loss during extended
period of bed rest and paralysis.
The animal models used to study disuse or unloading related bone adaptation
can be broadly classified as invasive or non-invasive. Invasive models include:
nerve resection, tenotomy and bone isolation. Both neurectomy and tenotomy
create disuse by disabling muscle-induced loading. Bone isolation involves
bypassing a part of bone as far as loading is concerned. Non-invasive models
include hindlimb suspension, limb casting/taping, space flight and botulinum toxin
induced muscle paralysis. The non-invasive disuse models, with the exception of
botulinum toxin induced muscle paralysis, do not inhibit muscle induced loading
of the bone and therefore do not interfere with physiological phenomenon asso-
ciated with low-level muscle activity.
Few studies have examined how aging influences unloading related bone loss.
Uhthoff et al. [66] reported the effects of 60 weeks of cast immobilizations in the
forelimb of young adult (1–3 year) and old (7–8 years) dogs. Bone loss was
greater in trabecular compared to cortical bone, and at distal (metacarpals) com-
pared to proximal (humerus) sites. This pattern was similar in adult and old dogs,
and the magnitude of bone loss was similar in the two age groups although the
young adult dogs lost bone by reduced periosteal expansion whereas older dogs
lost bone primarily by endosteal resorption. In slight contradiction, the same

authors reported that ‘‘older dogs appear to lose less bone than young adult dogs
during the same period of immobilization’’ [67]. Perrien et al. subjected young
(6 month) and old (32 month) male rats to 2 weeks of hindlimb unloading to study
changes at the proximal tibia [68]. Trabecular bone volume was significantly
decreased by unloading in young rats; old rats had markedly less trabecular bone
than young rats but this was not diminished further by unloading. On the other
hand, cortical bone mineral density was decreased and cortical porosity increased
by unloading in old but not young rats. Thus, unloading had negative effects on the
skeleton of young and old rats, although it primarily affected trabecular bone in
young animals and cortical bone in old animals.
6 Discussion
Our understanding of skeletal mechano-biology has come a long way from the early
observations by Roux and Wolff, largely due to animal studies that have identified
the loading parameters that most influence adaptation. However, the general rules
regarding bone adaptation are predominantly based on the observations in young
animals, and are predominantly based on cortical outcomes. It is not clear how
aging influences the dependence of bone adaptation on various loading parameters
such as strain magnitude, strain rate, loading frequency, etc. For example,
Srinivasan et al. [53] reported that old mice did not exhibit the dose response to
increased strain magnitude that young mice did, whereas we found that old mice did
exhibit a dose response [57]. Development of treatment strategies that rely on
modifying the mechanical environment of skeletal tissues for a favorable outcome
will require a better understanding of the interactions between mechanical loading
and the many biological factors that change with aging.
Improved understanding of the cellular/molecular level biophysical and bio-
chemical events involved with the sensing, transduction and response of bone cells
to mechanical stimuli, and data on how aging influences these events would
greatly facilitate the development of aforementioned therapies. For instance it has
been suggested that a deficit in the number of osteoblasts (the effector cell)
accounts for age-related loss of mechano-responsiveness. However, it is not known
if the sensing and transduction mechanisms are active and performing to their full
potential. It might be futile to focus our efforts on developing therapies to over-
come the age-related osteoblast deficit if the sensing and transduction mechanisms
(most likely related to osteocytes) are impaired. It should also be noted that aging
is often accompanied by various pathologies with potential to influence mechan-
ical loading/unloading related bone adaptation, such as diabetes and hypertension.
Studies to explore how such pathologies and aging together influence bone
mechano-responsiveness will prove to be quite challenging and it will likely take
years to understand such interactions.
We propose a few guidelines regarding design and interpretation of experiments

to study the influence of aging on loading/unloading induced bone adaptation.
First, especially in the case of older human subjects or animals that undergo
normal age-related bone loss, intrinsic loading may prevent bone loss rather than
stimulating bone accrual. In these settings, to correctly identify the mechanism
responsible for maintenance of bone health one has to study the temporal changes
at a skeletal site of interest. Temporal changes can be studied either by inclusion of
both baseline and age-matched control groups or by longitudinal in vivo microCT
or by dynamic histomorphometry measurements. Comparison of bone structure
and strength with sedentary controls at the end of the training/loading period does
not provide insight into the mechanism (bone accrual versus prevention of bone
resorption) responsible for the observed differences. Second, consideration should
be given to selection of appropriate age of the animals. To address the issue of
aging-related changes in bone adaptation the animal ages should be selected so as
to represent different stages of the lifespan (Table 1). As a rule, ages should be
considered relative to median survival age (the typical measure of longevity of
a species). Median lifespan for mice is approximately 24 months, for rats
30 months, and for humans 75 years. Of course these depend on gender, envi-
ronmental factors (e.g., diet) and genetic background (e.g., rat or mouse strain). As
noted above, many studies of ‘‘aging’’ do not include old animals! Studies com-
paring young to mature animals are relevant to bone accrual (peak bone mass),
whereas studies comparing mature to old animals are relevant to age-related bone
loss. Third, regarding the choice of loading modality in rodents there appears to be
a consensus toward use of either axial hindlimb (tibial) compression or forelimb
(ulnar) compression. Four- or three-point bending should be avoided due to
aforementioned issues of periosteal contact pressure. Tibial cantilever bending is a
valid model, but in our hands it is more difficult than tibial or ulnar compression.
An important distinction between axial compression versus cantilever bending is
that the former imposes loading in a direction similar to that for habitual loading,
whereas the latter imposes loading in a non-habitual direction. Either might be
suitable depending on the question being asked.
In summary, bone retains the ability to adapt in response to altered loading
environment at old age provided the induced strain level is sufficient. This
observation is supported by studies that have employed both intrinsic and extrinsic
loading modalities. However, the magnitude of the response of the aged skeleton may
be less than the young skeleton, and/or the threshold to trigger a response may be
greater. Given the relatively few studies that directly compare young, mature and old
animals, there remain many unanswered questions about aging and skeletal
mechanoresponsiveness and thus many opportunities to contribute answers.
Acknowledgments We thank Blaine Christiansen and Nilsson Holguin for their thoughtful
reviews of this chapter. We gratefully acknowledge support from the U.S. National Institutes of
Health NIH/NIAMS R01AR047867 and R21AR054371.
References
1. Martin, R.B., Burr, D.B., Sharkey, N.A.: Skeletal Tissue Mechanics. Springer, New York
(1998)
2. Carter, D.R.: Mechanical loading histories and cortical bone remodeling. Calcif. Tiss. Int. 36,
S19–S24 (1984)
3. Frost, H.M.: Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal. Anat. Record 219, 1–9 (1987)
4. Qin, Y.X., Rubin, C.T., McLeod, K.J.: Nonlinear dependence of loading intensity and cycle
number in the maintenance of bone mass and morphology. J. Orthop. Res. 16, 482–489
(1998)
5. Turner, C.H.: Three rules for bone adaptation to mechanical stimuli. Bone 23, 399–407
(1998)
6. Robling, A.G., Castillo, A.B., Turner, C.H.: Biomechanical and molecular regulation of bone
7. Busse, B., Djonic, D., Milovanovic, P., Hahn, M., Puschel, K., Ritchie, R.O., et al.: Decrease
in the osteocyte lacunar density accompanied by hypermineralized lacunar occlusion reveals
failure and delay of remodeling in aged human bone. Aging Cell 9, 1065–1075 (2010)
8. Mullender, M.G., van der Meer, D.D., Huiskes, R., Lips, P.: Osteocyte density changes in
aging and osteoporosis. Bone 18, 109–113 (1996)
9. Vashishth, D., Verborgt, O., Divine, G., Schaffler, M.B., Fyhrie, D.P.: Decline in osteocyte
lacunar density in human cortical bone is associated with accumulation of microcracks with
age. Bone 26, 375–380 (2000)
10. Perrini, S., Laviola, L., Carreira, M.C., Cignarelli, A., Natalicchio, A., Giorgino, F.: The
GH/IGF1 axis and signaling pathways in the muscle and bone: mechanisms underlying age-
related skeletal muscle wasting and osteoporosis. J. Endocrinol. 205, 201–210 (2010)
11. Walsh, M.C., Hunter, G.R., Livingstone, M.B.: Sarcopenia in premenopausal and
postmenopausal women with osteopenia, osteoporosis and normal bone mineral density.
Osteoporos. Int. 17, 61–67 (2006)
12. Osteoporosis Prevention, Diagnosis, and Therapy.: NIH Consensus Statement, 1–45 (2000)
13. Reginster, J.Y., Burlet, N.: Osteoporosis: a still increasing prevalence. Bone 38, S4–S9 (2006)
14. Rosen, C.J.: Clinical practice. Postmenopausal osteoporosis. N. Engl. J. Med. 353, 595–603
(2005)
15. Shane, E., Burr, D., Ebeling, P.R., Abrahamsen, B., Adler, R.A., Brown, T.D., et al.: 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 (2010)
16. Khosla, S., Westendorf, J.J., Oursler, M.J.: Building bone to reverse osteoporosis and repair
fractures. J. Clin. Invest. 118, 421–428 (2008)
17. Flurkey, K., Currer, J.M., Harrison, D.E.: Mouse models in aging research. In: Fox, J.G.,
Barthold, S.W., Davisson, M.T., Newcomer, C.E., Quimby, F.W., Smith, A.L. (eds.) The
Mouse in Biomedical Research, 2nd edn, pp. 637–672. Academic Press, Burlington (2007)
18. Robling, A.G., Burr, D.B., Turner, C.H.: Skeletal loading in animals. J. Musculoskelet.
Neuronal Interact. 1, 249–262 (2001)
19. Fritton, S.P., Rubin, C.T.: In vivo measurement of bone deformations using strain gauges.
In: Cowin, S.C. (ed.) Bone Mechanics Handbook, 2nd edn, pp. 8-1–8-41. CRC Press, Boca
Raton (2001)
20. Sztefek, P., Vanleene, M., Olsson, R., Collinson, R., Pitsillides, A.A., Shefelbine, S.: Using
digital image correlation to determine bone surface strains during loading and after
adaptation of the mouse tibia. J. Biomech. 43, 599–605 (2010)
21. Gross, T.S., Edwards, J.L., McLeod, K.J., Rubin, C.T.: Strain gradients correlate with sites of
periosteal bone formation. J. Bone Miner. Res. 12, 982–988 (1997)
22. Gross, T.S., Srinivasan, S., Liu, C.C., Clemens, T.L., Bain, S.D.: Noninvasive loading of the
murine tibia: an in vivo model for the study of mechanotransduction. J. Bone Miner. Res. 17,
493–501 (2002)
23. Kotha, S.P., Hsieh, Y.-F., Strigel, R.M., Muller, R., Silva, M.J.: Experimental and finite
element analysis of the rat ulnar loading model—correlations between strain and bone
formation following fatigue loading. J. Biomech. 37, 541–548 (2004)
24. Silva, M.J., Brodt, M.D., Hucker, W.J.: Finite element analysis of the mouse tibia: estimating
endocortical strain during three-point bending in SAMP6 osteoporotic mice. Anat. Rec.
A. Discov. Mol. Cell. Evol. Biol. 283A, 380–390 (2005)
25. Mosley, J.R., March, B.M., Lynch, J., Lanyon, L.E.: Strain magnitude related changes in
whole bone architecture in growing rats. Bone 20, 191–198 (1997)
26. Rabkin, B.A., Szivek, J.A., Schonfeld, J.E., Halloran, B.P.: Long-term measurement of bone
strain in vivo: the rat tibia. J. Biomed. Mater. Res. 58, 277–281 (2001)
27. Indrekvam, K., Husby, O.S., Gjerdet, N.R., Engester, L.B., Langeland, N.: Age-dependent
mechanical properties of rat femur. Measured in vivo and in vitro. Acta Orthop. Scand. 62,
248–252 (1991)
28. Keller, T.S., Spengler, D.M.: Regulation of bone stress and strain in the immature and mature
rat femur. J. Biomech. 22, 1115–1127 (1989)
29. Raab, D.M., Smith, E.L., Crenshaw, T.D., Thomas, D.P.: Bone mechanical properties after
exercise training in young and old rats. J. Appl. Physiol. 68, 130–134 (1990)
30. Leppanen, O.V., Sievanen, H., Jokihaara, J., Pajamaki, I., Kannus, P., Jarvinen, T.L.:
Pathogenesis of age-related osteoporosis: impaired mechano-responsiveness of bone is not
the culprit. PLoS ONE 3, e2540 (2008)
31. Bennell, K.L., Khan, K.M., Warmington, S., Forwood, M.R., Coleman, B.D., Bennett, M.B.,
et al.: Age does not influence the bone response to treadmill exercise in female rats. Med. Sci.
Sports Exerc. 34, 1958–1965 (2002)
32. Umemura, Y., Ishiko, T., Tsujimoto, H., Miura, H., Mokushi, N., Suzuki, H.: Effects of jump
training on bone hypertrophy in young and old rats. Int. J. Sports Med. 16, 364–367 (1995)
33. Jarvinen, T.L., Pajamaki, I., Sievanen, H., Vuohelainen, T., Tuukkanen, J., Jarvinen, M.,
et al.: Femoral neck response to exercise and subsequent deconditioning in young and adult
rats. J. Bone Miner. Res. 18, 1292–1299 (2003)
34. Hoshi, A., Watanabe, H., Chiba, M., Inaba, Y.: Effects of exercise at different ages on bone
density and mechanical properties of femoral bone of aged mice. Tohoku J. Exp. Med. 185,
15–24 (1998)
35. Buhl, K.M., Jacobs, C.R., Turner, R.T., Evans, G.L., Farrell, P.A., Donahue, H.J.: Aged bone
displays an increased responsiveness to low-intensity resistance exercise. J. Appl. Physiol.
90, 1359–1364 (2001)
36. Honda, A., Sogo, N., Nagasawa, S., Kato, T., Umemura, Y.: Bones benefits gained by jump
training are preserved after detraining in young and adult rats. J. Appl. Physiol. 105, 849–853
(2008)
37. Rubin, C.T., Lanyon, L.E.: Regulation of bone formation by applied dynamic loads. J. Bone
Jt. Surg. [Am] 66, 397–402 (1984)
38. Lanyon, L.E., Rubin, C.T.: Static vs dynamic loads as an influence on bone remodelling.
J. Biomech. 17, 897–905 (1984)
39. Chambers, T.J., Evans, M., Gardner, T.N.: Turner- Smith A, Chow JWM. Induction of bone
formation in rat tail vertebrae by mechanical loading. Bone Miner. 20, 167–178 (1993)
40. Kroeber M, Unglaub F, Guehring T, Nerlich A, Hadi T, Lotz J, et al.: Effects of controlled
dynamic disc distraction on degenerated intervertebral discs: an in vivo study on the rabbit
lumbar spine model. Spine (Phila Pa 1976) 2005; 30:181–187
41. Rubin, C.T., Bain, S.D., McCleod, K.J.: Suppression of osteogenic response in the aging
skeleton. Calcif. Tiss. Int. 50, 306–313 (1992)
42. Turner, C.H., Akhter, M.P., Raab, D.M., Kimmel, D.B., Recker, R.R.: A noninvasive, in vivo
model for studying strain adaptive bone modeling. Bone 12, 73–79 (1991)
43. Akhter, M.P., Cullen, D.M., Pedersen, E.A., Kimmel, D.B., Recker, R.R.: Bone response to
in vivo mechanical loading in two breeds of mice. Calcif. Tiss. Int. 63, 442–449 (1998)
44. Turner, C.H., Forwood, M.R., Rho, J.Y., Yoshikawa, T.: Mechanical loading thresholds for
lamellar and woven bone formation. J. Bone Min. Res. 9, 87–97 (1994)
45. Turner, C.H., Takano, Y., Owan, I.: Aging changes mechanical loading thresholds for bone
formation in rats. J. Bone Miner. Res. 10, 1544–1549 (1995)
46. Kesavan, C., Mohan, S., Oberholtzer, S., Wergedal, J.E., Baylink, D.J.: Mechanical loading-
induced gene expression and BMD changes are different in two inbred mouse strains. J. Appl.
Physiol. 99, 1951–1957 (2005)
47. Jilka, R.L., Weinstein, R.S., Takahashi, K., Parfitt, A.M., Manolagas, S.C.: Linkage of
decreased bone mass with impaired osteoblastogenesis in a murine model of accelerated
senescence. J. Clin. Invest. 97, 1732–1740 (1996)
48. Silva, M.J., Brodt, M.B., Ettner, S.L.: Long bones from the senescence accelerated mouse
SAMP6 have increased size but reduced whole-bone strength and resistance to fracture.
J. Bone Miner. Res. 17, 1597–1603 (2002)
49. Silva, M.J., Brodt, M.D., Ko, M., Abu-Amer, Y.: Impaired marrow osteogenesis is associated
with reduced endocortical bone formation but does not impair periosteal bone formation in
long bones of SAMP6 mice. J. Bone Miner. Res. 20, 419–427 (2005)
50. Matsushita, M., Tsuboyama, T., Kasai, R., Okumura, H., Yamamuro, T., Higuchi, K., et al.:
Age-related changes in bone mass in the senescence-accelerated mouse (SAM). SAM-R/3 and
SAM-P/6 as new murine models for senile osteoporosis. Am. J. Pathol. 125, 276–283 (1986)
51. Silva, M.J., Brodt, M.D.: Mechanical stimulation of bone formation is normal in the SAMP6
mouse. Calcif. Tissue Int. 82, 489–497 (2008)
52. Silva MJ, Brodt MD.: Bone biomechanics and mechanobiology in the SAMP6 mouse. In:
Takeda T, (ed.) The Senescence-Accelerated Mouse (SAM): Achievements and Future
Directions, Elsevier; London (2011) (estimated)
53. Srinivasan, S., Agans, S.C., King, K.A., Moy, N.Y., Poliachik, S.L., Gross, T.S.: Enabling bone
formation in the aged skeleton via rest-inserted mechanical loading. Bone 33, 946–955 (2003)
54. Fritton, J.C., Myers, E.R., Wright, T.M., van der Meulen, M.C.: Loading induces site-specific
increases in mineral content assessed by microcomputed tomography of the mouse tibia.
Bone 36, 1030–1038 (2005)
55. De Souza, R.L., Matsuura, M., Eckstein, F., Rawlinson, S.C., Lanyon, L.E., Pitsillides, A.A.:
Non-invasive axial loading of mouse tibiae increases cortical bone formation and modifies
trabecular organization: a new model to study cortical and cancellous compartments in a
single loaded element. Bone 37, 810–818 (2005)
56. Christiansen, B.A., Bayly, P.V., Silva, M.J.: Constrained tibial vibration in mice: a method
for studying the effects of vibrational loading of bone. J. Biomech. Eng. 130, 044502 (2008)
57. Brodt, M.D., Silva, M.J.: Aged mice have enhanced endocortical response and normal
periosteal response compared to young-adult mice following 1 week of axial tibial
compression. J. Bone Miner. Res. 25, 2006–2015 (2010)
58. Silva MJ, Brodt MD, Lynch MA, Stephens AL, Wood DJ, Civitelli R. (2012) Tibial Loading
Increases Osteogenic Gene Expression and Cortical Bone Volume in Mature and Middle-
Aged Mice. PLoS ONE 7(4), e34980. doi:10.1371/journal.pone.0034980
59. Lynch, M.E., Main, R.P., Xu, Q., Walsh, D.J., Schaffler, M.B., Wright, T.M., et al.:
Cancellous bone adaptation to tibial compression is not sex dependent in growing mice.
J. Appl. Physiol. 109, 685–691 (2010)
60. Lynch, M.E., Main, R.P., Xu, Q., Schmicker, T.L., Schaffler, M.B., Wright, T.M., et al.:
Tibial compression is anabolic in the adult mouse skeleton despite reduced responsiveness
with aging. Bone 49, 439–446 (2011)
61. Ozcivici, E., Luu, Y.K., Adler, B., Qin, Y.X., Rubin, J., Judex, S., et al.: Mechanical signals
as anabolic agents in bone. Nat. Rev. Rheumatol. 6, 50–59 (2010)
62. Huang, R.P., Rubin, C.T., McLeod, K.J.: Changes in postural muscle dynamics as a function
of age. J. Gerontol. A Biol. Sci. Med. Sci. 54, B352–B357 (1999)
63. Judex, S., Donahue, L.R., Rubin, C.: Genetic predisposition to low bone mass is paralleled by
an enhanced sensitivity to signals anabolic to the skeleton. Faseb J 16, 1260–1262 (2002)
64. Xie, L., Rubin, C., Judex, S.: Enhancement of the adolescent murine musculoskeletal system
using low-level mechanical vibrations. J. Appl. Physiol. 104, 1056–1062 (2008)
65. Lynch, M.A., Brodt, M.D., Silva, M.J.: Skeletal effects of whole-body vibration in adult and
aged mice. J. Orthop. Res. 28, 241–247 (2010)
66. Uhthoff HK, Sekaly G, Jaworski ZF. Effect of long-term nontraumatic immobilization on
metaphyseal spongiosa in young adult and old beagle dogs. Clin Orthop Relat Res
1985:278–283
67. Jaworski ZF, Liskova-Kiar M, Uhthoff HK. Effect of long-term immobilisation on the pattern
of bone loss in older dogs. J Bone Joint Surg Br 1980; 62-B:104-10
68. Perrien, D.S., Akel, N.S., Dupont-Versteegden, E.E., Skinner, R.A., Siegel, E.R., Suva, L.J.,
et al.: Aging alters the skeletal response to disuse in the rat. Am. J. Physiol. Regul. Integr.
Comp. Physiol. 292, R988–R996 (2007)
69. Brodt, M.D., Ellis, C.B., Silva, M.J.: Growing C57Bl/6 mice increase whole bone mechanical
properties by increasing geometric and material properties. J. Bone Miner. Res. 14,
2159–2166 (1999)
70. Harrison, D.E.: Baseline life span data: twelve strains of commonly used laboratory mice.
In: 2009
71. Holloszy, J.O.: Exercise increases average longevity of female rats despite increased food
intake and no growth retardation. J Gerontol 48, B97–B100 (1993)
72. Jee, W.S., Yao, W.: Overview: animal models of osteopenia and osteoporosis.
J. Musculoskelet. Neuronal Interact. 1, 193–207 (2001)
Skeletal Mechanoresponsiveness:
Effects of Sex Hormones
Katherine M. Melville, Natalie H. Kelly

and Marjolein C. H. van der Meulen
Abstract Sex hormones regulate bone mass, and their age-associated decline
contributes to bone loss seen clinically with menopause and aging. Mechanical
loading in surgical models of hormone deficiency has been examined extensively
as a therapy to overcome the decreased bone mass associated with sex hormone
deficiency. Exercise and controlled loading can overcome cancellous bone loss
following ovariectomy and orchidectomy in rodent models. In addition, several
signaling pathways associated with skeletal mechanotransduction have recently
been shown to be regulated by sex hormones or, more specifically, their receptors.
Deletion of hormone cellular receptors (estrogen receptors a and b, and androgen
receptor) in mice suggests a critical role for estrogen in the response of bone tissue
to mechanical stimuli. In this chapter we review the literature on skeletal adap-
tation to mechanical loading in surgical and genetic rodent models of sex hormone
deficiency.
1 Introduction
Mechanical loading is a critical regulator of skeletal mass and structure starting

during embryonic development and continuing into senescence. During growth
and development, additional mechanical stimuli can further enhance already-
increasing bone mass whereas reduced loading slows or inhibits growth. In the
K. M. Melville N. H. Kelly M. C. H. van der Meulen

College of Engineering, Cornell University, Ithaca, NY 14853, USA
M. C. H. van der Meulen (&)
Research Division, Hospital for Special Surgery, New York, NY 10021, USA
e-mail: mcv3@cornell.edu

DOI: 10.1007/8415_2012_135
Published Online: 10 May 2012
218 K. M. Melville et al.
adult, increased mechanical loading is anabolic and enhances bone mass, whereas
reduced mechanical loading results in bone loss. The effects of loading are
modulated by a number of other factors that individually influence bone mass.
Sex hormones are also critical regulators of skeletal growth in males and
females during periods of increasing and decreasing skeletal mass. At least 50% of
adult bone peak mass is accrued during puberty, a stage of rapid bone growth [1].
In the healthy adult skeleton, both estrogens (the primary female sex hormones)
and androgens (the primary male sex hormones) positively impact bone remod-
eling and the maintenance of bone mass, primarily by suppressing resorption and
bone remodeling [2–5]. In women, estrogen deficiency following menopause
contributes to the rapid decline in bone mass and decreased skeletal structural
capacity that can lead to osteoporosis and fracture [5, 6]. Declines in sex hormones
with age in males are more gradual but produce similar effects [7, 8].
The tissue-level effects of estrogen and estrogen withdrawal on bone mass in
the presence of mechanical stimuli are well documented [5, 9, 10]. In preclinical
models using skeletally mature animals, hormone deficiency produces cancellous
bone loss initially with subsequent cortical bone loss. While estrogen deficiency
uniformly increases bone turnover, the loss of cancellous bone mass is not uniform
and may in fact be related to the complex distribution of mechanical stimuli in the
skeleton [10, 11].
Identifying the role of sex hormones on the mechanoresponsiveness of the
skeleton is critical to understanding aging and developing therapies for age- and
hormone-related bone loss. Reduced responsiveness to exercise has been reported
in female rodents compared to males and could reflect not only hormonal but also
growth factor differences [12, 13]. From a practical perspective, mechanical
loading is a candidate anabolic stimulus to overcome and treat hormone-defi-
ciency-induced bone loss. A variety of loading approaches have been examined in
animal models to counteract hormone deficiency with variable success [14–17].
Mechanistically, several cellular pathways for bone mechanotransduction are
regulated by interactions between estrogen and cellular estrogen receptors (ERs)
present in bone cells [18–20] (Fig. 1). The responses of estrogen and androgen
receptor deficient mouse models to controlled skeletal loading have been examined
to elucidate the signaling pathways and adaptive mechanisms. A great deal
remains to be learned about the bone anabolic and anti-resorptive actions induced
by ERs; progress has been limited by available mouse genetic models to isolate
specific contributions and skeletal mechanotransduction approaches to study the
effects of mechanical stimuli in vivo.
The skeletal response to surgically induced hormone deficiency is well
established in a variety of animal models [21]. In preclinical models, hormone
deficiency can be induced by surgical removal of the gonads in both males
(orchidectomy, ORX) and females (ovariectomy, OVX). In adult rodents ORX and
OVX both produce cancellous bone loss initially. Until recently the adult OVX rat
was the most commonly used model; however, the focus has shifted to mouse
models due to the ability to characterize and manipulate the mouse genome.
Transgenic technology has led to the creation of knockout (KO) mouse models
Skeletal Mechanoresponsiveness 219
Fig. 1 Schematic showing

pathways of estrogen and
androgen signaling in the
skeleton with mechanical
loading. (E2: estradiol, the
predominant estrogen; T:
testosterone, the predominant
androgen; ERa: estrogen
receptor-alpha; ERb:
estrogen receptor-beta; AR:
androgen receptor)
of estrogen receptor-a (ERaKO), estrogen receptor-b (ERbKO), and androgen

receptors (ARKO) [22].
Loading can be combined with both surgical and genetic models of hormone
deficiency. As described in the previous chapter, increased in vivo loading can
be achieved through exercise (intrinsic) and direct (extrinsic) skeletal loading
(see ‘‘The Effect of Aging on Skeletal Mechanoresponsiveness: Animal Studies’’,
Kotiya and Silva). A limited number of studies have examined reduced loading
combined with hormone deficiency. Most of our knowledge comes from
increasing the in vivo loading in hormone-deficient rats and more recently mice;
large animal models of combined exercise and hormone effects are limited [23].
In this chapter we will focus on mechanotransduction in rodent models of sex
hormone deficiency.
2 Hormone-Deficiency Induced Models of Osteoporosis

and In Vivo Loading
Sex hormone deficiency results in bone loss and can be induced in preclinical
studies by surgical removal of the gonads to simulate the natural decreases in
hormone production with aging in humans. Preclinical models of surgically
induced hormone deficiency in rodents demonstrate the key features of bone loss
seen clinically [24]. Measures to not only counteract but also inhibit this bone loss
and the associated morbidities such as fractures have been studied extensively.
While pharmacological treatments are currently the clinical standard [25],
biophysical stimuli such as mechanical loading could be used to treat and possibly
also inhibit bone loss resulting from age related sex hormone deficiency.
Exercise has been used extensively in preclinical studies as a treatment for
hormone-induced bone loss. Treadmill running is the most commonly used
exercise approach in rodent studies [10, 12, 16, 17, 26–28]. Other exercise routines
studied include tower/ladder climbing [29–31] and jumping [15, 32]. Whole body
vibration also perturbs the mechanical environment of the skeleton and has been
used in several studies [33–38].
The mechanical environment of the skeleton during exercise is complex
and difficult to quantify; therefore, mechanical parameters such as peak load
magnitude cannot be directly related to adaptation. The mechanical environment
during whole body vibration is similarly difficult to determine. Methods that
directly load the skeleton in vivo, such as tibial four-point bending [39] or axial
tibial compression [40], allow the applied loading to be controlled and quantified.
These extrinsic loading protocols reduce the confounding effect of body mass and
load level differences present in exercise studies. Controlled loading methods have
been used to examine the skeletal response to loading in the absence of hormones
[14, 41]. Here we focus on mechanical loading from exercise and direct loading
approaches applied to hormone-deficient rodent models.
The majority of studies examining osteoporosis induced by surgical sex
hormone deficiency have been performed in the ovariectomized (OVX) rat.
Appropriate controls involve sham surgery, subjecting the animals to similar
treatment and handling. OVX rats gain considerable fat mass following surgery
[42], and outweigh sham surgical controls even when pair-feeding is performed.
OVX of the skeletally mature rat (8–9 months of age) is a well-characterized
model that captures key features of cancellous and cortical bone changes seen in
humans [43]. Growing rats can be used to study skeletal changes with hormone
deficiency and have greater bone loss with OVX than adults, but changes in bone
mass occur through different mechanisms than are present in postmenopausal
women [42]. The other primary rodent model, the mouse, is less well-established
as a surgical model of osteoporosis. Bone changes with hormone deficiency vary
by mouse strain [44]. However, the mouse is a powerful tool for studying signaling
and genetic regulation of bone mass [45, 46], as will be described later.
2.1 Hormone-Deficiency and Exercise
Hormone deficiency due to surgical removal of the gonads decreases cancellous

and cortical bone volumes. Treadmill exercise increases bone mass following OVX,
but consistent recovery of cancellous bone measures to the levels of sham-operated
controls occurs only in growing animals [12, 16, 17]. Rats that underwent OVX at
weaning (3 weeks) and subsequently exercised by treadmill for 16 weeks were more
responsive to mechanical loading at the femoral neck than sham-operated animals
[12]. The OVX-induced reduction in femoral neck strength was compensated
by exercise and even increased above sham levels. In adolescent (3-months at time of
OVX) female rats, OVX decreased cancellous bone volume in the distal femur by
40–50% compared to sham-operated animals [16, 17]. Treadmill exercise partially
rescued the bone volume lost due to OVX, but the levels remained below sham levels
[16, 17]. However, exercise did completely counteract the decreased mechanical
strength of the femoral neck and tibial shaft with OVX, returning strength levels to
those of sham rats [16, 17]. Importantly, in these studies measures of mass and
volume were not accurate surrogates for bone strength. The effect on cancellous bone
volume fraction was similar in adult (5-months) rats exercised on a treadmill for
30 mins daily, but whole bone strength was not measured [26]. Skeletally mature
(C8-months) rats also respond to exercise following OVX, but the effects are more
variable, with both partial and complete rescue of bone loss reported [47–50].
The response to loading following OVX is age-dependent; exercise in juvenile
rats restores bone to equal or greater levels than those of sham animals, but
increases in older rats are insufficient to reach sham levels. In most exercise
experiments, only the OVX group responded to exercise, whereas the sham
animals did not. However, in young rats OVX generally increased body mass over
control levels and partially protected the animals from osteopenia; therefore, the
increased body mass loading with OVX may contribute to the skeletal changes
seen with running [51].
Other forms of intrinsic skeletal loading by exercise, such as tower or ladder
climbing, can also recover the bone loss associated with hormone deficiency in rats.
Tower climbing for 3-months in mature (12-months) rats recovered bone loss by
thickening the remaining trabeculae, without changing trabecular separation [15].
OVX in combination with exercise led to similar bone mass and strength as sedentary
sham-surgery rats. Exercise prevented OVX-induced cortical and cancellous bone
loss by depressing the elevated bone turnover following OVX [15]. Adolescent
(3-months) OVX rats that were trained to climb a ladder for 12 weeks had increased
tibial BMD, tissue mineral density, and bending strength compared to intact
and OVX sedentary controls, and similar values to intact exercised rats [32].
High impact exercise caused by jumping strengthened bones in OVX rats with
relatively few numbers of jumps. These protocols are effective in both sham and OVX
animals. Jump training for 2-months increased tibial bone mass, cortical area and
whole bone strength in both young (3-months) and adult (9-months) rats with only ten
daily jumps [29, 30]. In young rats, jump training significantly increased serum
osteocalcin in both OVX and sham operated animals, hinting at increased osteoblast
activity [30]. In adult rats, bone mass and strength of OVX rats increased with
exercise to approximately the same levels as sham-exercised rats [29]. Similar results
have been reported for different jump training protocols. When four bouts of ten
jumps were performed every 48 h, ash weight, BMD, and bone strength increased in
the tibia of OVX rats compared to sedentary controls [31].
The anabolic effect of exercise with hormone deficiency occurs through both
increased bone formation and decreased bone resorption, as well as combined
effects. In the rat, estrogen deficiency creates increased bone turnover and bone
resorption [24, 42]. Increased numbers of osteoclasts following OVX were reduced
by exercise [17, 26]. Exercise plus OVX also increased osteoblast activity com-
pared to OVX alone [26]. In growing male rats, ORX reduced osteoblast-lined
surface and osteoclast number after eight weeks, and exercise returned these
indices to sham levels, normalizing bone turnover with ORX [27].
The majority of exercise studies have examined OVX; fewer studies have
examined the role of exercise in bone loss following ORX in male rodents. In
growing male rats, exercise did not protect against cancellous bone loss following
ORX. When young (3 month) male rats were exercised on a treadmill, cancellous
bone volume was decreased at both time points in ORX rats with and without
exercise [27]. ORX also decreased the mechanical strength of the femoral neck,
and exercise did not prevent this reduction. Thus, for growing rats exercised at a
similar speed, duration and intensity, the skeletons of OVX female rats may be
more sensitive to the anabolic effects of exercise than those of ORX male rats.
Shorter daily exercise durations and moderate intensity may be most beneficial
in counteracting bone loss following estrogen deficiency [26, 17, 52]. Moderate
exercise attenuated OVX-induced cancellous bone loss and increased cortical bone
mass in adult (6-months) rats [26]. Exercise for 30 min/day increased cancellous
bone volume, with decreased resorption and increased osteoblastic activity,
compared to the sedentary OVX group but was unable to return to the level of
sham controls [26]. Cortical bone area was not different between the OVX and
sham groups, but increased with 30 min of daily exercise compared to OVX alone.
Bone measures were not different in OVX rats with 60 min/day of exercise
compared to OVX without exercise. When multiple intensities were compared,
running at moderate intensity (10–12 m/min) was more effective at restoring bone
to sham levels than faster speeds (18 m/min) for the same duration [17, 53].
Normal, growth-related gains in ash weight were lower in OVX alone and vig-
orously exercised OVX animals, but similar in moderately exercised OVX and
sham animals. Femoral neck strength decreased with OVX and returned to control
levels with moderate intensity running. OVX decreased trabecular bone volume by
52% compared to sham, and moderate exercise reduced this loss to 30%, while
more intense exercise reduced the loss to 40%. The exercise-induced increases in
cancellous bone and bending strength of the humerus differed with exercise
intensity and were greatest with moderate intensity running. Clearly, intensity and
periodicity of exercise can have major impacts on how bone responds to exercise.
Animal studies have focused on administering estrogen to counteract hormonal
loss seen with OVX. The administration of estrogen or phytoestrogens prevented
bone loss associated with OVX and had an additive effect with exercise [10, 28,
49, 54]. Bone loss was prevented with pharmacological replacement of estrogen
(17b-estradiol) in growing rats that underwent OVX and subsequent sciatic
neurectomy [10]. Similar results were found in adult OVX rats when 17b-estradiol
was combined with treadmill exercise [49]. While combined treatment did not
restore tibial cancellous bone mass to control levels, exercise and estrogen
replacement together were additive and significantly more effective than either
treatment alone. Similarly, combining Genistein, an isoflavone that interacts with
estrogen receptors, with moderate treadmill exercise was additive in preventing
bone loss in ORX mice [28]. In growing (2-months) mice, 17b-estradiol alone
completely prevented OVX- and ORX-induced cancellous bone loss in the femur
[28, 54].
2.2 Hormone Deficiency and Direct Skeletal Loading
In contrast to exercise models of skeletal loading, the mechanical stimulus applied

to the skeleton is quantifiable and accurately controlled with direct (extrinsic)
loading methods. While a large number of studies have combined OVX with
exercise, only limited data are available for direct loading [14, 41]. The response
of the tibia to direct skeletal loading by four-point bending of the diaphysis and by
axial compression of the whole hindlimb has been examined with sex hormone
deficient animals. The target strain range in the cortex with loading is gener-
ally *1200–1300 le, corresponding to locomotory strains measured across a
range of species [55].
Direct loading of the tibia inhibits bone loss with hormone deficiency. Tibial
four-point bending of adult (6-months) OVX and sham rats increased cortical area
through equivalent dynamic measures of bone formation [14]. Similar loads were
applied to OVX and sham rats (31.4 N), producing mean strains of *1300 le in
both groups. The cortical bone response to loading was not altered with OVX and
the absence of endogenous estrogen. Likewise, 6 weeks of cyclic tibial com-
pression, producing 1200 le at the tibial midshaft of mice, inhibited bone loss after
ORX and maintained absolute bone mass at age-matched sham control levels in
growing (2-months) mice [41]. ORX decreased cancellous bone volume fraction
and trabecular number, increased trabecular separation, and did not change tra-
becular thickness. Metaphyseal cancellous bone volume was greater with loading
than in contralateral nonloaded tibia, due to trabecular thickening with loading.
Loading was associated with greater mineral apposition rates and smaller percent
mineralizing surfaces. Cortical mass decline was less severe than cancellous bone
loss with ORX, but diaphyseal cortical properties were reduced compared to sham
surgery mice. In cortical bone, loading increased BMC and maximum moment of
inertia similarly in both sham and ORX mice. In vivo tibial compression increased
cancellous and cortical bone mass in osteopenic OVX adult female mice after
6 weeks of loading (unpublished data). In OVX mice, cancellous bone mass in
loaded limbs exhibited a bimodal distribution with time due to the competing
effects of loading and estrogen deficiency. Adult (6-months old) C57Bl/6 mice
were osteopenic, and estrogen deficiency did not further reduce cortical bone
mass. The ability of the skeleton to adapt to mechanical loading was unaltered with
OVX. As in exercise studies, the ability to form bone with direct skeletal
loading does not require endogenous estrogen and is similar in intact and hormone-
deficient mice.
3 Genetic Models of Sex Hormone Deficiency
Estrogen and testosterone are signaling hormones that act via estrogen receptors
(ERs) and androgen receptor (AR), respectively, located in the cytosol (Fig. 1). Sex
hormones are involved in reproductive signaling, but their receptors are also
expressed in bone cells (osteoblasts, osteoclasts, and osteocytes) at levels around ten
times lower than that of reproductive tissues [56]. The two known estrogen receptors
are estrogen receptor alpha (ERa) and estrogen receptor beta (ERb). When estrogen
interacts with an ER, the receptor translocates to the nucleus, phosphorylates,
dimerizes, and mediates gene transcription via classical estrogen response elements
or via direct interaction with DNA binding sites to increase cell number and activity
[57]. Androgen actions are mediated through the AR, which then dimerizes and can
regulate gene transcription through androgen response elements (AREs) although
non-genomic activities have also been implicated [58, 59].
Both ERs and ARs have important implications in bone maintenance (Vico and
Vanacker). In 1994, a man was discovered who had an inactivating point mutation
in the ERa gene [60]. He presented with open growth plates and severe osteopo-
rosis, suggesting that ERa plays an important role in bone maturation and
homeostasis. Although not the focus of this chapter, a number of in vitro studies
subsequently demonstrated the importance of ERa in the anabolic response of bone
cells to mechanical strain [61–65]. (See ‘‘Bone Cell Mechanoresponsiveness’’,
Genetos and Jacobs.) Normally, mechanical loading enhances cell proliferation in
vitro. When the effects of ERa are blocked using selective estrogen receptor
modulators, strain-induced proliferation of osteoblast-like cells derived from rat
long bones was reduced [62]. Similarly, osteoblast-like cells from ERaKO mice
proliferated less in response to mechanical loading compared to wild type (WT)
cells [64, 66, 67]. Furthermore, ERa can be activated not only by estrogen, but also
by mechanical strain alone or in combination with estrogen [68]. In vitro, the effects
of loading on bone cells from ERbKO mice are opposite from those observed in
cells from ERaKO mice. Mechanical stimulation doubled proliferation of cultured
osteoblast-like cells from ERbKO mice compared to WT cells [64, 66].
In the following sections we will review the roles of ERs and AR in bone
mechanotransduction based on in vivo studies examining ERaKO, ERbKO and
ARKO mice. The skeletal phenotype resulting from the receptor deletion will be
presented first, followed by results from controlled loading studies.
3.1 Deletion of Estrogen Receptor Alpha
The first ERaKO mice were generated in 1993 by an insertional gene mutation, but
since then improved ERaKO mice have been generated to minimize ERa protein
detection and gene expression [69–72]. Absence of ERa affects both reproductive
tissue and bone phenotypes in female mice. Body mass and body fat are both
increased due to decreased estrogen action, similar to what results when estrogen
is depleted by OVX in mice [73]. Deletion of ERa reduces uterine weight com-
pared to wild type (WT) mice [74]. Long bones are shorter in female ERaKO mice
[64, 74] and correlated with decreased serum levels of IGF-1 [73, 74]. Despite
shorter bone lengths, the ERaKO female skeleton is more mineralized in some
locations. In the proximal tibia, trabecular BMD was increased in growing female
ERaKO mice and even more so in young adult mice; cortical BMD and thickness
in the tibia were also increased in both growing and skeletally mature ERaKO
females compared to WT [73]. In the ulna, cortical area was increased, but cortical
stiffness was similar between genotypes [64]. However, overall BMC was
unchanged in ERaKO females compared to WT mice [73]. ERa plays a critical
role in bone properties of female mice, but higher levels of estrogen and testos-
terone along with lower levels of osteocalcin found in these knockout mice may
indicate compensatory mechanisms including estrogen acting via ERb [69, 73].
As in female mice, in male mice the absence of ERa adversely affects both
cortical and cancellous bone from puberty onwards, characterized by shorter bone
lengths, smaller cortical area and decreased trabecular density compared to WT
males [75, 76]. Despite bone phenotypic differences found in male ERaKO mice,
their response to mechanical loading has not been extensively studied.
Non-invasive in vivo loading has given further insight to the role of ERa in
bone mechanotransduction. Under normal circumstances, in vivo mouse ulnar and
tibial loading increases bone formation and BMC of loaded limbs compared to
contralateral control limbs of mice [40, 77–79]. In ERaKO female mice, this
response is severely attenuated. The first reported in vivo loading of female
ERaKO mice loaded the ulna of 20–24 week-old female mice [67]. After 2 weeks,
cortical area in the midshaft of the ulna of WT mice increased 8%, but in the
ERaKO mice, cortical area increased only 2.4%. In a subsequent ulnar loading
experiment, cortical area increased three-fold less in ERaKO mice compared to
WT mice after two weeks of loading [64]. The increased area was primarily due to
periosteal expansion (80%) with a smaller contribution from endosteal bone for-
mation. In addition, MAR and MS increased less with mechanical loading in
ERaKO than in WT.
In vivo loading studies show that female ERaKO mice did not exhibit the same
anabolic response to controlled mechanical loading as WT mice. In a unique
examination of the genetic profile of bone cells from loaded and non-loaded tibiae
from ERaKO, OVX, and WT female mice, the right tibiae were loaded for 60
cycles at 2 Hz, with peak strains at the midshaft of 1300 le [80]. At 3, 8, 12 and
24 h after loading, loaded limbs of WT mice had greater differential response to
loading than either OVX WT or ERaKO mice when assessed by microarray and
qRT-PCR. For example, at the 3-h time point, only 26 genes were differentially
regulated in the ERaKO mice, compared with 642 for WT mice. These data give
insight to ERa’s critical role in mechanotransduction signaling and gene expres-
sion. ERa has been linked to a number of mechanotransduction signaling
pathways, including Wnt/b-catenin, IGF-1, and PTH [19].
3.2 Deletion of Estrogen Receptor b
ERbKO mice were generated later than ERaKO mice, and have revealed a
different role for this receptor than for ERa [70, 81]. Just as in ERaKO mice, adult
female ERbKO mice are heavier with higher fat mass than their WT littermates
[82, 83]. Since ERb is found on osteocytes and osteoblasts as well as reproductive
organs, lack of ERb affects bones, but reports have varied [84].
Long bones of growing and adult female ERbKO mice have been reported to be
longer [73, 83] and similar to WT values [64, 83, 85]. Interestingly, during growth
the skeleton of ERbKO female mice appears to be adversely affected, but during
aging the trend reverses. In growing ERbKO mice, volumetric BMD (vBMD) of
the lumbar spine and distal femur was decreased compared to WT [82]. Mice
usually develop osteopenia in trabecular-rich regions with age [86], but ERbKO
mice are protected from this bone loss. At one year, trabecular density in both tibia
and femur was higher in ERbKO females than WT by pQCT [87]. This protective
effect is even more evident in cortical bone; cortical area, BMC and cortical
thickness were increased in the tibia and femur of 1-year-old female ERbKO mice
compared to WT [87].
Male ERbKO bones, in contrast, exhibit few abnormalities. Bone length, total
BMC and cortical density and area were similar in tibiae and femora from ERbKO
and WT mice [87]. Interestingly, male ERbKO male body mass and some skeletal
phenotypes are similar to female WT, suggesting a ‘‘feminization’’ of ERbKO
male bones [87].
Aside from the skeletal phenotypes, other systemic effects were present in ERbKO
mice. Reports of altered hormone levels have varied. The initial female ERbKO mouse
models did not have altered estrogen serum levels [73, 87]. Subsequent female ERbKO
mouse models showed increased serum estrogen levels [82]. Interestingly, at 1 year of
age, ERa mRNA levels were two-fold higher in ERbKO female mice than WT
females, suggesting that ERb may suppress effects of ERa in the genome, or that ERa
compensates for the lack of ERb through a yet unknown mechanism [87].
While the in vitro response of bone cells to loading is increased in the absence
of ERb [20], the in vivo loading data for the ulnae of ERbKO mice are conflicting.
The inconsistent data may reflect differences in loading protocols or the different
genetic mutation of the ERb gene. In a mouse model with an insertional mutation
in the ERb gene, the loading-induced increase in cortical area and MAR in the
ulna was less in adult (5-months-old) female ERbKO mice compared to WT
controls [64]. In contrast, using adult (4-months-old) mice with an exon 3 deletion
in the ERb gene, MAR and BFR increased more in response to loading in the ulna
in female ERbKO mice compared to WT [82]. In the same study, male ERbKO
and WT mice responded similarly to in vivo loading [82]. Therefore, the role of
ERb is more apparent in females than males, both through phenotype and through
response to mechanical loading.
3.3 Deletion of Androgen Receptor
Generating ARKO mice is a genetic challenge because the AR gene is present on

the X chromosome and required for male fertility, making female generations of
ARKO mice difficult to create and therefore less studied [88]. A number of male
ARKO mouse models have been generated; this chapter will only focus on the
results in bone tissue [88, 89].
Male ARKO mice differed phenotypically from their WT controls. During
growth body mass is lower in male ARKO mice than WT males [90]. However,
after 10 weeks of age, ARKO male mice rapidly increase mass to surpass the WT
males and eventually become obese [91]. At 8 weeks, serum testosterone levels
were lower in male ARKO mice compared to WT [90, 92]. Cortical and trabecular
bone are both severely affected by lack of AR at 8 weeks of age. In the femur,
cross sectional area, cortical area, and cortical thickness were all lower than in
WT [93]. In both the femur and tibia, trabecular bone volume was decreased to
osteopenic levels [92, 93]. These cortical and cancellous trends continued into
adulthood [94]. Interestingly, bone formation rate in both the tibia and femur was
increased, but it was accompanied by increased resorption rates, which exceeded
the bone formation rates and thus resulted in osteopenia [92, 93].
Although bone resorption rates in growing male ARKO mice exceed bone
formation rates in WT mice, in vivo mechanical loading can still stimulate
bone formation [95]. The left ulnae of 5-months-old male ARKO and WT mice
were subjected to 40 cycles of loading with peak strains of 1560–1740 le [95].
Stiffness was similar for both genotypes at the start of the experiment. After 2
weeks, periosteal bone formation rates increased with loading in both genotypes,
but the change was significantly greater in ARKO mice, possibly due to a
disruption in the Wnt/b-catenin pathway, known to play a role in bone’s
mechanoresponsiveness [96]. SOST, a gene that codes for sclerostin, an inhibitor
of the Wnt/b-catenin pathway, had decreased expression after mechanical loading
in WT mice, but more so in ARKO mice. In contrast, the number of sclerostin-
positive osteocytes decreased in ARKO mice after mechanical loading, but not in
WT. These data suggest a role for AR in the Wnt/b-catenin signaling pathway in
addition to ERa’s hypothesized role.
To further study AR’s role in bone adaptation, ARKO mice have been subjected
to both increased and decreased physical activity with different results in cortical
versus cancellous bone. In an exercise study, ARKO male mice were subjected to
voluntary exercise from age 1 to 4-months [94]. Bone turnover parameters were
lower in sedentary than exercised ARKO mice, as expected. Furthermore, tra-
becular bone mineral density, bone volume, and trabecular number were greater in
the tibiae of exercised ARKO compared to sedentary mice. However, running did
not affect cortical bone gain in exercised ARKO mice compared to sedentary mice.
In contrast, the combination of AR deficiency and hind limb unloading affected
both trabecular and cortical bone in ARKO mice, demonstrating the importance of
AR in bone maintenance especially under disuse conditions [97]. Following tail
suspension for up to 2 weeks, 2-months old male ARKO mice showed rapid bone
loss, with a 70% decrease in trabecular volume in the hindlimbs [97]. Trabecular
bone volume decreased in both ARKO and WT mice, but the loss was greater in
ARKO mice. Cortical thickness and bone area were decreased in ARKO mice after
tail suspension, but not in WT. These exercise studies indicate that AR deletion
does not prevent bone adaptation to mechanical loading, but that decreased
physical activity in the absence of AR may be more detrimental to bone than in
mice with an intact AR.
4 Summary and Conclusions
The role of sex hormones in bone mechanotransduction has been studied in vivo
through two different approaches in rodents: surgical models of sex hormone
deficiency and genetic knockout models. The former removes circulating
bioavailable estrogen but leaves receptors and signaling pathways intact; the latter
approach targets specific components of the signaling pathway, primarily the sex
hormone receptors to date. Surgical models combined with either exercise or
applied loading demonstrate that loading can overcome the bone loss resulting
from sex hormone withdrawal. This work has focused primarily on cancellous
bone loss, but cortical effects are also present. Results comparing different loading
protocols show that not all interventions are equally successful. Particular aspects
of the loading protocol may be critical to evoking a response including moderate
intensity and duration of exercise. The responses to loading include both inhibiting
the resorption activated by sex hormone deficiency and activating bone formation.
An important limitation of this work is the lack of Haversian remodeling in
rodents, which may produce different adaptive responses in larger animal models.
From these studies, circulating estrogen does not appear to be required for the
anabolic skeletal response to exercise or applied mechanical loading.
Mechanistically, genetic models that remove estrogen or androgen receptors
provide a tremendous opportunity to understand the signaling contributions of sex
hormones in mechanotransduction. Based on genome-wide removal of estrogen
receptors in the mouse, the ERs play important roles in both cancellous and
cortical mechanotransduction. In experiments with altered mechanical loading,
the absence of ERs is more severe than the absence of circulating estrogen. This
result likely reflects the multiple signal pathways to which ERs contribute in
addition to classical estrogen signaling [19, 98, 99]. Their precise roles will be
elucidated as more in vivo data become available for individual cell types and
specific signaling pathways. Early data suggest a role for ERa in multiple
key pathways that control bone remodeling and adaptation. Future tissue-
specific knockouts aimed at isolating the effects of estrogen on individual cell
populations will provide valuable information about the role of sex hormones
in mechanotransduction.
Acknowledgments This work was supported by the National Institutes of Health (R01-
AG028664, R01-AR053571) and the National Science Foundation (GRFs to KMM and NHK).
References
1. Moro, M., van der Meulen, M.C., Kiratli, B.J., Marcus, R., Bachrach, L.K., Carter, D.R.:
Body mass is the primary determinant of midfemoral bone acquisition during adolescent
growth. Bone 19(5), 519–526 (1996). doi: 10.1007/s00774-010-0176-1. S8756328296002633
[pii]
2. Eastell, R.: Role of oestrogen in the regulation of bone turnover at the menarche.
J. Endocrinol. 185(2), 223–234 (2005). doi:10.1677/joe.1.06059.185/2/223 [pii]
3. Manolagas, S.C., Bellido, T., Jilka, R.L.: Sex steroids, cytokines and the bone marrow: New
concepts on the pathogenesis of osteoporosis. Ciba Found Symp. 191:187–196. discussion
197–202 (1995)
4. Riggs, B.L.: The mechanisms of estrogen regulation of bone resorption. J. Clin. Invest.
106(10), 1203–1204 (2000). doi:10.1172/JCI11468
5. Riggs, B.L.: Endocrine causes of age-related bone loss and osteoporosis. Novartis Found
Symp. 242:247–259. discussion 260–244 (2002)
6. Seeman, E.: Pathogenesis of bone fragility in women and men. Lancet 359(9320), 1841–1850
(2002). doi:10.1016/S0140-6736(02)08706-8
7. Callewaert, F., Boonen, S., Vanderschueren, D.: Sex steroids and the male skeleton: A tale
of two hormones. Trends Endocrinol. Metab. 21(2), 89–95 (2010). doi:10.1016/
j.tem.2009.09.002
8. Falahati-Nini, A., Riggs, B.L., Atkinson, E.J., O’Fallon, W.M., Eastell, R., Khosla, S.:
Relative contributions of testosterone and estrogen in regulating bone resorption and
formation in normal elderly men. J. Clin. Invest. 106(12), 1553–1560 (2000). doi:10.1172/
JCI10942
9. Cavolina, J.M., Evans, G.L., Harris, S.A., Zhang, M., Westerlind, K.C., Turner, R.T.: The
effects of orbital spaceflight on bone histomorphometry and messenger ribonucleic acid
levels for bone matrix proteins and skeletal signaling peptides in ovariectomized growing
rats. Endocrinology 138(4), 1567–1576 (1997)
10. Westerlind, K.C., Wronski, T.J., Ritman, E.L., Luo, Z.P., An, K.N., Bell, N.H., Turner, R.T.:
Estrogen regulates the rate of bone turnover but bone balance in ovariectomized rats is
modulated by prevailing mechanical strain. Proc. Natl. Acad. Sci. U S A 94(8), 4199–4204
(1997)
11. Bagi, C.M., Miller, S.C.: Comparison of osteopenic changes in cancellous bone induced by
ovariectomy and/or immobilization in adult rats. Anat. Rec. 239(3), 243–254 (1994).
doi:10.1002/ar.1092390303
12. Jarvinen, T.L., Kannus, P., Pajamaki, I., Vuohelainen, T., Tuukkanen, J., Jarvinen, M.,
Sievanen, H.: Estrogen deposits extra mineral into bones of female rats in puberty, but
simultaneously seems to suppress the responsiveness of female skeleton to mechanical
loading. Bone 32(6), 642–651 (2003). S8756328203001005 [pii]
13. Wallace, J.M., Rajachar, R.M., Allen, M.R., Bloomfield, S.A., Robey, P.G., Young, M.F.,
Kohn, D.H.: Exercise-induced changes in the cortical bone of growing mice are bone- and
gender-specific. Bone 40(4), 1120–1127 (2007). S8756-3282(06)00912-4 [pii]
14. Hagino, H., Raab, D.M., Kimmel, D.B., Akhter, M.P., Recker, R.R.: Effect of ovariectomy on
bone response to in vivo external loading. J. Bone. Miner. Res. 8(3), 347–357 (1993).
doi:10.1002/jbmr.5650080312
15. Notomi, T., Okimoto, N., Okazaki, Y., Nakamura, T., Suzuki, M.: Tower climbing exercise
started 3 months after ovariectomy recovers bone strength of the femur and lumbar vertebrae
in aged osteopenic rats. J. Bone. Miner. Res. 18(1), 140–149 (2003). doi:10.1359/jbmr.
2003.18.1.140
16. Peng, Z., Tuukkanen, J., Vaananen, H.K.: Exercise can provide protection against bone loss
and prevent the decrease in mechanical strength of femoral neck in ovariectomized rats.
J. Bone. Miner. Res. 9(10), 1559–1564 (1994). doi:10.1002/jbmr.5650091008
17. Peng, Z.Q., Vaananen, H.K., Tuukkanen, J.: Ovariectomy-induced bone loss can be affected
by different intensities of treadmill running exercise in rats. Calcif. Tissue. Int. 60(5),
441–448 (1997)
18. Lanyon, L., Armstrong, V., Ong, D., Zaman, G., Price, J.: Is estrogen receptor alpha key to
controlling bones’ resistance to fracture? J. Endocrinol. 182(2), 183–191 (2004)
19. Price, J.S., Sugiyama, T., Galea, G.L., Meakin, L.B., Sunters, A., Lanyon, L.E.: Role of
endocrine and paracrine factors in the adaptation of bone to mechanical loading. Curr.
Osteoporos. Rep. 9(2), 76–82 (2011). doi:10.1007/s11914-011-0050-7
20. Saxon, L.K., Turner, C.H.: Estrogen receptor beta: the antimechanostat? Bone 36(2),
185–192 (2005). doi:1016/j.bone.2004.08.003
21. Iwaniec, I., Turner, R.: Animal models for osteoporosis. In: Marcus, R., Feldman, D., Nelson, D.A.,
Rosen, C.J. (eds) Osteoporosis, vol. 2, 3rd edn. Academic Press, pp 985–1009 (2008)
22. Vico, L., Vanacker, J.M.: Sex hormones and their receptors in bone homeostasis: insights
from genetically modified mouse models. Osteoporos. Int. 21(3), 365–372 (2010).
doi:10.1007/s00198-009-0963-5
23. Devlin, M.J., Lieberman, D.E.: Variation in estradiol level affects cortical bone growth
in response to mechanical loading in sheep. J. Exp. Biol. 210(Pt 4), 602–613 (2007).
doi:10.1242/jeb.02675
24. Wronski, T.J., Lowry, P.L., Walsh, C.C., Ignaszewski, L.A.: Skeletal alterations
in ovariectomized rats. Calcif. Tissue Int. 37(3), 324–328 (1985)
25. Rachner, T.D., Khosla, S., Hofbauer, L.C.: Osteoporosis: now and the future. Lancet
377(9773), 1276–1287 (2011). doi:10.1016/S0140-6736(10)62349-5
26. Iwamoto, J., Takeda, T., Ichimura, S.: Effects of moderate intensity exercise on tibial bone mass
in mature ovariectomized rats: bone histomorphometry study. Keio J Med 47(3), 162–167
(1998)
27. Tuukkanen, J., Peng, Z., Vaananen, H.K.: Effect of running exercise on the bone loss induced
by orchidectomy in the rat. Calcif. Tissue. Int. 55(1), 33–37 (1994)
28. Wu, J., Wang, X.X., Chiba, H., Higuchi, M., Takasaki, M., Ohta, A., Ishimi, Y.: Combined
intervention of exercise and genistein prevented androgen deficiency-induced bone loss in
mice. J. Appl. Physiol. 94(1), 335–342 (2003). doi:10.1152/japplphysiol.00498.2002
29. Honda, A., Sogo, N., Nagasawa, S., Shimizu, T., Umemura, Y.: High-impact exercise
strengthens bone in osteopenic ovariectomized rats with the same outcome as sham rats.
J. Appl. Physiol. 95(3), 1032–1037 (2003). doi:10.1152/japplphysiol.00781.2002. 00781.2002
[pii]
30. Honda, A., Umemura, Y., Nagasawa, S.: Effect of high-impact and low-repetition training
on bones in ovariectomized rats. J. Bone Miner. Res. 16(9), 1688–1693 (2001). doi:10.1359/
jbmr.2001.16.9.1688
31. Renno, A.C., Silveira Gomes, A.R., Nascimento, R.B., Salvini, T., Parizoto, N.: Effects of a
progressive loading exercise program on the bone and skeletal muscle properties of female
osteopenic rats. Exp. Gerontol. 42(6), 517–522 (2007). doi:10.1016/j.exger.2006.11.014
32. Shiguemoto, G.E., Prestes, J., Leite, R.D., Pereira, G.B., Pontes, C.L., D’Avila, F.V., Botero, J.P.,
Baldissera, V., Nonaka, K.O., Selistre-de-Araujo, H.S., Perez, S.E.: Effects of resistance training
on matrix metalloproteinase-2 activity and biomechanical and physical properties of bone
in ovariectomized and intact rats. Scand. J. Med. Sci. Sports (2011). doi:10.1111/j.1600-
0838.2010.01284.x
33. Brouwers, J.E., van Rietbergen, B., Ito, K., Huiskes, R.: Effects of vibration treatment on
tibial bone of ovariectomized rats analyzed by in vivo micro-CT. J. Orthop. Res. 28(1), 62–69
(2010). doi:10.1002/jor.20951
34. Judex, S., Lei, X., Han, D., Rubin, C.: Low-magnitude mechanical signals that stimulate bone
formation in the ovariectomized rat are dependent on the applied frequency but not on the
strain magnitude. J. Biomech. 40(6), 1333–1339 (2007). doi:10.1016/j.jbiomech.2006.05.014
35. Oxlund, B.S., Ortoft, G., Andreassen, T.T., Oxlund, H.: Low-intensity, high-frequency
vibration appears to prevent the decrease in strength of the femur and tibia associated with
ovariectomy of adult rats. Bone 32(1), 69–77 (2003). S875632820200916X [pii]
36. Rubinacci, A., Marenzana, M., Cavani, F., Colasante, F., Villa, I., Willnecker, J., Moro, G.L.,
Spreafico, L.P., Ferretti, M., Guidobono, F., Marotti, G.: Ovariectomy sensitizes rat cortical
bone to whole-body vibration. Calcif. Tissue Int. 82(4), 316–326 (2008). doi:10.1007/
s00223-008-9115-8
37. Sehmisch, S., Galal, R., Kolios, L., Tezval, M., Dullin, C., Zimmer, S., Stuermer, K.M.,
Stuermer, E.K.: Effects of low-magnitude, high-frequency mechanical stimulation in the rat
osteopenia model. Osteoporos Int. 20(12), 1999–2008 (2009). doi:10.1007/s00198-009-0892-3
38. Tezval, M., Biblis, M., Sehmisch, S., Schmelz, U., Kolios, L., Rack, T., Stuermer, K.M.,
Stuermer, E.K.: Improvement of femoral bone quality after low-magnitude, high-frequency
mechanical stimulation in the ovariectomized rat as an osteopenia model. Calcif. Tissue Int.
88(1), 33–40 (2011). doi:10.1007/s00223-010-9423-7
39. Turner, C.H., Akhter, M.P., Raab, D.M., Kimmel, D.B., Recker, R.R.: A noninvasive, in vivo
model for studying strain adaptive bone modeling. Bone 12(2), 73–79 (1991)
40. Fritton, J.C., Myers, E.R., Wright, T.M., van der Meulen, M.C.: Loading induces site-specific
increases in mineral content assessed by microcomputed tomography of the mouse tibia.
Bone 36(6), 1030–1038 (2005). doi:10.1016/j.bone.2005.02.013
41. Fritton, J.C., Myers, E.R., Wright, T.M., van der Meulen, M.C.: Bone mass is preserved and
cancellous architecture altered due to cyclic loading of the mouse tibia after orchidectomy.
J. Bone Miner. Res. 23(5), 663–671 (2008). doi:10.1359/jbmr.080104
42. Thompson, D.D., Simmons, H.A., Pirie, C.M., Ke, H.Z.: FDA guidelines and animal models
for osteoporosis. Bone 17(4 Suppl), 125S–133S (1995)
43. Jee, W.S., Yao, W.: Overview: Animal models of osteopenia and osteoporosis. J. Musculoskelet.
Neuronal. Interact. 1(3), 193–207 (2001)
44. Bouxsein, M.L., Myers, K.S., Shultz, K.L., Donahue, L.R., Rosen, C.J., Beamer, W.G.:
Ovariectomy-induced bone loss varies among inbred strains of mice. J. Bone Miner. Res.
20(7), 1085–1092 (2005). doi:10.1359/JBMR.050307
45. Robling, A.G., Li, J., Shultz, K.L., Beamer, W.G., Turner, C.H.: Evidence for a skeletal
mechanosensitivity gene on mouse chromosome 4. FASEB J. 17(2), 324–326 (2003).
doi:10.1096/fj.02-0393fje02-0393fje
46. Rosen, C.J., Beamer, W.G., Donahue, L.R.: Defining the genetics of osteoporosis: Using the
mouse to understand man. Osteoporos. Int. 12(10), 803–810 (2001)
47. Barengolts, E.I., Curry, D.J., Bapna, M.S., Kukreja, S.C.: Effects of endurance exercise on
bone mass and mechanical properties in intact and ovariectomized rats. J. Bone Miner. Res.
8(8), 937–942 (1993). doi:10.1002/jbmr.5650080806
48. Pohlman, R.L., Darby, L.A., Lechner, A.J.: Morphometry and calcium contents in appendicular
and axial bones of exercised ovariectomized rats. Am. J. Physiol. 248(1 Pt 2), 12–17 (1985)
49. Yeh, J.K., Liu, C.C., Aloia, J.F.: Additive effect of treadmill exercise and 17 beta-estradiol
replacement on prevention of tibial bone loss in adult ovariectomized rat. J. Bone Miner. Res.
8(6), 677–683 (1993). doi:10.1002/jbmr.5650080605
50. Barengolts, E.I., Curry, D.J., Bapna, M.S., Kukreja, S.C.: Effects of two non-endurance
exercise protocols on established bone loss in ovariectomized adult rats. Calcif. Tissue Int.
52(3), 239–243 (1993)
51. Wronski, T.J., Schenck, P.A., Cintron, M., Walsh, C.C.: Effect of body weight on osteopenia
in ovariectomized rats. Calcif. Tissue Int. 40(3), 155–159 (1987)
52. Iwamoto, J., Takeda, T., Ichimura, S.: Effects of exercise on bone mineral density in
mature osteopenic rats. J. Bone Miner. Res. 13(8), 1308–1317 (1998). doi:10.1359/
jbmr.1998.13.8.1308
53. Iwamoto, J., Takeda, T., Ichimura, S.: Effect of exercise on tibial and lumbar vertebral bone
mass in mature osteopenic rats: bone histomorphometry study. J. Orthop. Sci. 3(5), 257–263
(1998)
54. Wu, J., Wang, X.X., Takasaki, M., Ohta, A., Higuchi, M., Ishimi, Y.: Cooperative effects of
exercise training and genistein administration on bone mass in ovariectomized mice. J. Bone
Miner. Res. 16(10), 1829–1836 (2001). doi:10.1359/jbmr.2001.16.10.1829
55. Rubin, C.T., Lanyon, L.E.: Dynamic strain similarity in vertebrates; an alternative to
allometric limb bone scaling. J. Theor. Biol. 107(2), 321–327 (1984)
56. Manolagas, S.C., Kousteni, S., Jilka, R.L.: Sex steroids and bone. Recent Prog. Horm. Res.
57, 385–409 (2002)
57. Tsai, M.J., O’Malley, B.W.: Molecular mechanisms of action of steroid/thyroid receptor
superfamily members. Annu. Rev. Biochem. 63, 451–486 (1994). doi:10.1146/annurev.
bi.63.070194.002315
58. Heinlein, C.A., Chang, C.: The roles of androgen receptors and androgen-binding proteins in
nongenomic androgen actions. Mol. Endocrinol. 16(10), 2181–2187 (2002)
59. Kousteni, S., Bellido, T., Plotkin, L.I., O’Brien, C.A., Bodenner, D.L., Han, L., Han, K.,
DiGregorio, G.B., Katzenellenbogen, J.A., Katzenellenbogen, B.S., Roberson, P.K.,
Weinstein, R.S., Jilka, R.L., Manolagas, S.C.: Nongenotropic, sex-nonspecific signaling
through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell
104(5), 719–730 (2001). doi:S0092-8674(01)00268-9
60. Smith, E.P., Boyd, J., Frank, G.R., Takahashi, H., Cohen, R.M., Specker, B., Williams, T.C.,
Lubahn, D.B., Korach, K.S.: Estrogen resistance caused by a mutation in the estrogen-
receptor gene in a man. N. Engl. J. Med. 331(16), 1056–1061 (1994). doi:10.1056/
NEJM199410203311604
61. Cheng, M.Z., Rawlinson, S.C., Pitsillides, A.A., Zaman, G., Mohan, S., Baylink, D.J.,
Lanyon, L.E.: Human osteoblasts’ proliferative responses to strain and 17beta-estradiol are
mediated by the estrogen receptor and the receptor for insulin-like growth factor i. J. Bone
Miner. Res. 17(4), 593–602 (2002). doi:10.1359/jbmr.2002.17.4.593
62. Damien, E., Price, J.S., Lanyon, L.E.: The estrogen receptor’s involvement in osteoblasts’
adaptive response to mechanical strain. J. Bone Miner. Res. 13(8), 1275–1282 (1998).
doi:10.1359/jbmr.1998.13.8.1275
63. Ehrlich, P.J., Noble, B.S., Jessop, H.L., Stevens, H.Y., Mosley, J.R., Lanyon, L.E.: The effect
of in vivo mechanical loading on estrogen receptor alpha expression in rat ulnar osteocytes.
J. Bone Miner. Res. 17(9), 1646–1655 (2002). doi:10.1359/jbmr.2002.17.9.1646
64. Lee, K.C., Jessop, H., Suswillo, R., Zaman, G., Lanyon, L.E.: The adaptive response of bone
to mechanical loading in female transgenic mice is deficient in the absence of oestrogen
receptor-alpha and -beta. J. Endocrinol. 182(2), 193–201 (2004)
65. Zaman, G., Jessop, H.L., Muzylak, M., De Souza, R.L., Pitsillides, A.A., Price, J.S., Lanyon,
L.L.: Osteocytes use estrogen receptor a to respond to strain but their ERa content is regulated by
estrogen. J. Bone Miner. Res. 21(8), 1297–1306 (2006). doi:10.1359/jbmr.060504
66. Jessop, H.L., Suswillo, R.F., Rawlinson, S.C., Zaman, G., Lee, K., Das-Gupta, V., Pitsillides,
A.A., Lanyon, L.E.: Osteoblast-like cells from estrogen receptor alpha knockout mice have
deficient responses to mechanical strain. J. Bone Miner. Res. 19(6), 938–946 (2004). doi:10.1359/
jbmr.2004.19.6.938
67. Lee, K., Jessop, H., Suswillo, R., Zaman, G., Lanyon, L.: Endocrinology: bone adaptation requires
oestrogen receptor-alpha. Nature 424(6947), 389 (2003). doi:10.1038/424389a424389a
68. Jessop, H.L., Sjoberg, M., Cheng, M.Z., Zaman, G., Wheeler-Jones, C.P., Lanyon, L.E.:
Mechanical strain and estrogen activate estrogen receptor alpha in bone cells. J. Bone Miner.
Res. 16(6), 1045–1055 (2001). doi:10.1359/jbmr.2001.16.6.1045
69. Couse, J.F., Curtis, S.W., Washburn, T.F., Lindzey, J., Golding, T.S., Lubahn, D.B.,
Smithies, O., Korach, K.S.: Analysis of transcription and estrogen insensitivity in the female
mouse after targeted disruption of the estrogen receptor a gene. Mol. Endocrinol. 9(11),
1441–1454 (1995)
70. Dupont, S., Krust, A., Gansmuller, A., Dierich, A., Chambon, P., Mark, M.: Effect of single
and compound knockouts of estrogen receptors a (ERa) and b (ERb) on mouse reproductive
phenotypes. Development 127(19), 4277–4291 (2000)
71. Feng, Y., Manka, D., Wagner, K.U., Khan, S.A.: Estrogen receptor-a expression in the
mammary epithelium is required for ductal and alveolar morphogenesis in mice. Proc. Natl.
Acad. Sci. U S A 104(37), 14718–14723 (2007). doi:10.1073/pnas.0706933104
72. Lubahn, D.B., Moyer, J.S., Golding, T.S., Couse, J.F., Korach, K.S., Smithies, O.: Alteration
of reproductive function but not prenatal sexual development after insertional disruption of
the mouse estrogen receptor gene. Proc. Natl. Acad. Sci. U S A 90(23), 11162–11166 (1993)
73. Lindberg, M.K., Alatalo, S.L., Halleen, J.M., Mohan, S., Gustafsson, J.A., Ohlsson, C.:
Estrogen receptor specificity in the regulation of the skeleton in female mice. J. Endocrinol.
171(2), 229–236 (2001). JOE04348 [pii]
74. Vidal, O., Lindberg, M., Savendahl, L., Lubahn, D.B., Ritzen, E.M., Gustafsson, J.A., Ohlsson, C.:
Disproportional body growth in female estrogen receptor-alpha-inactivated mice. Biochem.
Biophys. Res. Commun. 265(2), 569–571 (1999). doi:10.1006/bbrc.1999.1711
75. Parikka, V., Peng, Z., Hentunen, T., Risteli, J., Elo, T., Vaananen, H.K., Harkonen, P.:
Estrogen responsiveness of bone formation in vitro and altered bone phenotype in aged
estrogen receptor-alpha-deficient male and female mice. Eur. J. Endocrinol. 152(2), 301–314
(2005). doi:10.1530/eje.1.01832
76. Vidal, O., Lindberg, M.K., Hollberg, K., Baylink, D.J., Andersson, G., Lubahn, D.B., Mohan, S.,
Gustafsson, J.A., Ohlsson, C.: Estrogen receptor specificity in the regulation of skeletal growth
and maturation in male mice. Proc. Natl. Acad. Sci. U S A 97(10), 5474–5479 (2000). 97/10/5474
[pii]
77. De Souza, R.L., Matsuura, M., Eckstein, F., Rawlinson, S.C., Lanyon, L.E., Pitsillides, A.A.:
Non-invasive axial loading of mouse tibiae increases cortical bone formation and modifies
trabecular organization: A new model to study cortical and cancellous compartments in a
single loaded element. Bone 37(6), 810–818 (2005). doi:10.1016/j.bone.2005.07.022
78. Gross, T.S., Srinivasan, S., Liu, C.C., Clemens, T.L., Bain, S.D.: Noninvasive loading of the
murine tibia: An in vivo model for the study of mechanotransduction. J. Bone Miner. Res.
17(3), 493–501 (2002). doi:10.1359/jbmr.2002.17.3.493
79. Lee, K.C., Maxwell, A., Lanyon, L.E.: Validation of a technique for studying functional
adaptation of the mouse ulna in response to mechanical loading. Bone 31(3), 407–412 (2002).
doi:S8756328202008426
80. Zaman, G., Saxon, L.K., Sunters, A., Hilton, H., Underhill, P., Williams, D., Price, J.S.,
Lanyon, L.E.: Loading-related regulation of gene expression in bone in the contexts of
estrogen deficiency, lack of estrogen receptor alpha and disuse. Bone 46(3), 628–642 (2010).
S8756-3282(09)01989-9 [pii]
81. Krege, J.H., Hodgin, J.B., Couse, J.F., Enmark, E., Warner, M., Mahler, J.F., Sar, M., Korach, K.S.,
Gustafsson, J.A., Smithies, O.: Generation and reproductive phenotypes of mice lacking estrogen
receptor beta. Proc. Natl. Acad. Sci. U S A 95(26), 15677–15682 (1998)
82. Saxon, L.K., Robling, A.G., Castillo, A.B., Mohan, S., Turner, C.H.: The skeletal
responsiveness to mechanical loading is enhanced in mice with a null mutation in estrogen
receptor-beta. Am. J. Physiol. Endocrinol. Metab. 293(2), 484–491 (2007). doi:10.1152/
ajpendo.00189.2007
83. Windahl, S.H., Vidal, O., Andersson, G., Gustafsson, J.A., Ohlsson, C.: Increased cortical
bone mineral content but unchanged trabecular bone mineral density in female ERb(-/-)
mice. J Clin Invest 104(7), 895–901 (1999). doi:10.1172/JCI6730
84. Onoe, Y., Miyaura, C., Ohta, H., Nozawa, S., Suda, T.: Expression of estrogen receptor beta
in rat bone. Endocrinology 138(10), 4509–4512 (1997)
85. Chagin, A.S., Lindberg, M.K., Andersson, N., Moverare, S., Gustafsson, J.A., Savendahl, L.,
Ohlsson, C.: Estrogen receptor-b inhibits skeletal growth and has the capacity to mediate
growth plate fusion in female mice. J. Bone Miner. Res. 19(1), 72–77 (2004). doi:10.1359/
JBMR.0301203
86. Silbermann, M., Weiss, A., Reznick, A.Z., Eilam, Y., Szydel, N., Gershon, D.: Age-related
trend for osteopenia in femurs of female c57bl/6 mice. Compr. Gerontol. A. 1(1), 45–51 (1987)
87. Windahl, S.H., Hollberg, K., Vidal, O., Gustafsson, J.A., Ohlsson, C., Andersson, G.: Female
estrogen receptor b-/- mice are partially protected against age-related trabecular bone loss.
J. Bone Miner. Res. 16(8), 1388–1398 (2001). doi:10.1359/jbmr.2001.16.8.1388
88. Kerkhofs, S., Denayer, S., Haelens, A., Claessens, F.: Androgen receptor knockout and
knock-in mouse models. J. Mol. Endocrinol. 42(1), 11–17 (2009). doi:10.1677/JME-08-0122
89. Sato, T., Kawano, H., Kato, S.: Study of androgen action in bone by analysis of androgen-
receptor deficient mice. J. Bone Miner. Metab. 20(6), 326–330 (2002). doi:10.1007/
s007740200047
90. Yeh, S., Tsai, M.Y., Xu, Q., Mu, X.M., Lardy, H., Huang, K.E., Lin, H., Yeh, S.D.,
Altuwaijri, S., Zhou, X., Xing, L., Boyce, B.F., Hung, M.C., Zhang, S., Gan, L., Chang, C.:
Generation and characterization of androgen receptor knockout (ARKO) mice: An in vivo
model for the study of androgen functions in selective tissues. Proc. Natl. Acad. Sci. U S A
99(21), 13498–13503 (2002). doi:10.1073/pnas.212474399212474399
91. Sato, T., Matsumoto, T., Yamada, T., Watanabe, T., Kawano, H., Kato, S.: Late onset of
obesity in male androgen receptor-deficient (ARKO) mice. Biochem. Biophys. Res.
Commun. 300(1), 167–171 (2003). doi:S0006291X02027742
92. Kawano, H., Sato, T., Yamada, T., Matsumoto, T., Sekine, K., Watanabe, T., Nakamura, T.,
Fukuda, T., Yoshimura, K., Yoshizawa, T., Aihara, K., Yamamoto, Y., Nakamichi, Y.,
Metzger, D., Chambon, P., Nakamura, K., Kawaguchi, H., Kato, S.: Suppressive function of
androgen receptor in bone resorption. Proc. Natl. Acad. Sci. U S A 100(16), 9416–9421
(2003). doi:10.1073/pnas.15335001001533500100
93. Venken, K., De Gendt, K., Boonen, S., Ophoff, J., Bouillon, R., Swinnen, J.V., Verhoeven, G.,
Vanderschueren, D.: Relative impact of androgen and estrogen receptor activation in the effects
of androgens on trabecular and cortical bone in growing male mice: A study in the androgen
receptor knockout mouse model. J. Bone Miner. Res. 21(4), 576–585 (2006). doi:10.1359/
jbmr.060103
94. Ophoff, J., Callewaert, F., Venken, K., De Gendt, K., Ohlsson, C., Gayan-Ramirez, G.,
Decramer, M., Boonen, S., Bouillon, R., Verhoeven, G., Vanderschueren, D.: Physical
activity in the androgen receptor knockout mouse: Evidence for reversal of androgen
deficiency on cancellous bone. Biochem. Biophys. Res. Commun. 378(1), 139–144 (2009).
doi:10.1016/j.bbrc.2008.11.016
95. Callewaert, F., Bakker, A., Schrooten, J., Van Meerbeek, B., Verhoeven, G., Boonen, S.,
Vanderschueren, D.: Androgen receptor disruption increases the osteogenic response to
mechanical loading in male mice. J. Bone Miner. Res. 25(1), 124–131 (2010). doi:10.1359/
jbmr.091001
96. Bonewald, L.F., Johnson, M.L.: Osteocytes, mechanosensing and wnt signaling. Bone 42(4),
606–615 (2008). doi:10.1016/jbone.200712.224
97. Saita, Y., Nakamura, T., Mizoguchi, F., Nakashima, K., Hemmi, H., Hayata, T., Ezura, Y.,
Kurosawa, H., Kato, S., Noda, M.: Combinatory effects of androgen receptor deficiency and
hind limb unloading on bone. Horm. Metab. Res. 41(11), 822–828 (2009). doi:10.1055/s-
0029-1231056
98. Lau, K.H., Kapur, S., Kesavan, C., Baylink, D.J.: Up-regulation of the wnt, estrogen receptor,
insulin-like growth factor-i, and bone morphogenetic protein pathways in C57BL/6J osteoblasts
as opposed to C3H/HeJ osteoblasts in part contributes to the differential anabolic response to
fluid shear. J. Biol. Chem. 281(14), 9576–9588 (2006). doi:10.1074/jbc.M509205200
99. Sunters, A., Armstrong, V.J., Zaman, G., Kypta, R.M., Kawano, Y., Lanyon, L.E., Price, J.S.:
Mechano-transduction in osteoblastic cells involves strain-regulated estrogen receptor alpha-
mediated control of insulin-like growth factor (IGF) I receptor sensitivity to ambient IGF, leading
to phosphatidylinositol 3-kinase/AKT-dependent wnt/LRP5 receptor-independent activation of
beta-catenin signaling. J. Biol. Chem. 285(12), 8743–8758 (2010). doi:M109.027086
Effects of Exercise and Physical
Interventions on Bone: Clinical Studies
Wendy M. Kohrt, Karen L. Villalon and Daniel W. Barry
Abstract Perhaps the best evidence that physical activity is essential for the
maintenance of bone mass and strength is the rapid and profound loss of bone
mineral that occurs during conditions of disuse, such as immobilization, bed rest,
and spaceflight. Physical activity throughout the lifespan has the potential to
reduce the risk for osteoporotic fracture by augmenting the development of peak
bone mass during childhood, maintaining bone mass during early adulthood, and
slowing the inevitable loss of bone mass in old age. However, the types and
amounts of physical activity needed to optimize skeletal integrity across the
lifespan and reduce osteoporotic fracture risk have not been precisely defined. This
chapter reviews the clinical evidence that physical activity is associated with
reduced fracture risk and that exercise training can increase or slow the decline in
bone mineral density (BMD) in adults. The clinical relevance of the key deter-
minants of the response of bone to mechanical loading that have evolved from
preclinical studies of animals (e.g., high strain magnitude, high strain rate, few
repetitions, unique strain distribution) is discussed. Novel factors that may influ-
ence the skeletal adaptation to exercise in humans are also discussed.
W. M. Kohrt (&) K. L. Villalon D. W. Barry

Department of Medicine, University of Colorado Denver,
Aurora, CO, USA
e-mail: wendy.kohrt@ucdenver.edu

DOI: 10.1007/8415_2011_91
Published Online: 4 August 2011
236 W. M. Kohrt et al.
1 Physical Activity and Fracture Prevention
There is limited, but encouraging, evidence that physical activity reduces the
incidence of osteoporotic fractures. Solid evidence from randomized controlled
trials is lacking, but prospective cohort and case–control studies suggest that
physical activity is associated with reduced fracture risk.
1.1 Randomized Controlled Trials
There have been no large randomized controlled trials (RCT) to determine whether
an exercise intervention reduces the incidence of osteoporotic fractures. However,
two small RCTs that evaluated the effects of exercise on BMD and other osteo-
porosis risk factors conducted long-term follow-up evaluations of fracture inci-
dence in study participants [50, 88]. Sinaki et al. randomized 50 postmenopausal
women to undergo 2 years of back strengthening exercise or no exercise and
evaluated the incidence of vertebral fractures 8 years after the completion of the
intervention [88]. There were 14 fractures among 322 vertebrae examined in
the control group (4.3%), compared with six fractures among 378 vertebrae in the
exercise group (1.6%; P = 0.03). Korpelainen et al. randomized 160 women aged
70–73 years to undergo 2.5 years of balance, leg strength, and impact exercises or
no exercise and evaluated the incidence of fractures after an average follow-up of
7 years [50]. The incident rate of any fracture was 0.05 per 1,000 person-years in
exercisers versus 0.08 per 1,000 person-years in controls [incidence rate ratio,
0.68; 95% confidence interval (CI), 0.34–1.32]. There were no hip fractures among
women in the exercise group and five among controls (P = 0.02). Because neither
of these studies reported on falls, it is not clear whether the potential anti-fracture
benefit of exercise is related to improvements in bone strength and/or a reduction
in falls.
1.2 Prospective Cohort and Case–Control Studies
A systematic review of the association of physical activity with fracture risk was
conducted in 2008 for the development of the 2008 Physical Activity Guidelines
for Americans [74]. One of the major conclusions from that review was that there
is an inverse association of physical activity with fracture risk (i.e., increased
physical activity, reduced fracture risk), particularly for hip fractures.
Both spine and hip fractures are of high clinical concern because of the related
morbidity and mortality. Few studies have evaluated the association of physical
activity and vertebral fracture risk and results have been mixed [27, 82, 84, 87, 88],
possibly because such fractures are more difficult to diagnose than hip fractures
Effects of Exercise and Physical Interventions on Bone 237
Robbins 2007
Michaelsson 2007
Kujala 2000
Hoidrup 2001
Prospective
Cohort Studies Gregson 2010
Gregg 1998
Feskanich 2002
Benetou 2010
Kanis 1999
Case-control Jaglal 1995
Studies Women
Farahmand 2000
Men
Boonyaratavej 2001
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Relative Risk of Hip Fracture
Most Active versus Least Active
Fig. 1 Point estimates of relative risk (±95% confidence intervals) of hip fracture from studies
that examined multiple levels of physical activity (most active group versus least active group)
(i.e., can be asymptomatic) and because there is a lack of consensus on the degree
of vertebral deformity that constitutes a fracture. A number of observational
studies have evaluated whether physical activity is associated with reduced risk for
hip fracture [7, 10, 21, 22, 27, 28, 32, 34, 37, 51, 65, 77]. In all of these studies, the
relative risk of hip fracture in the most active subgroup of the cohort was less than
in the least active subgroup, as evidenced by a relative risk ratio less than 1.0
(Fig. 1), but not all risk ratios were statistically significant. Based on these
observation studies, physical activity appears to reduce the risk for hip fracture in
both women and men.
It is difficult to determine the minimal physical activity exposure associated
with fracture protection from these studies because many used only categorical
levels of activity (e.g., low, medium, high). Among the studies that used quanti-
fiable metrics, the minimal levels of physical activity found to be significantly
associated with reduced fracture risk were: [9–14.9 MET-h/week (MET = met-
abolic equivalent, where 1 MET is an oxygen uptake of 3.5 mL/min/kg body
weight) of physical activity [22], [4 h/week of walking [22], [1290 kcal/week of
physical activity [27], and [1 h/week of physical activity [21, 37]. These levels of
physical activity were associated with relative reductions in hip fracture risk of
33–41%. Although these observational studies provide a guideline for the volume
of physical activity likely to be effective, they do not indicate whether it was the
frequency, duration, or intensity of the activity that influenced fracture risk. Two
studies that categorized physical activity by intensity (e.g., easy vs. normal vs.
brisk walking pace) found that higher-intensity physical activity was associated
with reduced fracture risk [22, 27]. Such observations are concordant with the
Adjusted Relative Hip Fracture Risk

1.2
1.0
0.8
0.6
0.4
Linear trend, p<0.001
0.2
<3 3-8.9 9-14.9 15-23.9 24+

Leisure Time Activity, MET.h/wk
1.2
1.0
0.8
0.6
0.4
Linear trend, p=0.02
0.2
<1 1 2-3 4+
Walking, h/wk
1.2
1.0
0.8
0.6
0.4
0.2
Easy Average Brisk/Very Brisk

Walking Pace
Fig. 2 Associations of hip fracture risk with leisure time physical activity (top panel), walking time
(middle panel), and walking pace (bottom panel) in postmenopausal women. Adapted from [22]
finding from preclinical studies that the magnitude of the loading force is a key
determinant of the adaptive response of bone [83].
The Nurses’ Health Study is a good example of how prospective cohort studies
have provided insights on the associations of physical activity with hip fracture
risk [22]. The findings of this study suggest that postmenopausal women should
accrue at least 9–14.9 MET.h of physical activity per week to reduce fracture risk
(Fig. 2, top panel) or at least 4 h/week of walking (Fig. 2, middle panel). The
results further suggest that a fast walking pace is more likely to reduce fracture risk
than a slow walking pace (Fig. 2, bottom panel).
2 Exercise and BMD
There have been numerous randomized controlled trials of the effects of exercise
training on BMD, which is a major determinant of fracture risk. Because many
of the intervention trials have been relatively small, it is not surprising that at
least 19 meta-analyses have been conducted to integrate the findings across
studies (Table 1) [8, 9, 38–45, 53, 58–62, 93, 95]. Only four of these meta-
analyses reported no significant benefit of exercise training on BMD at any
skeletal region [8, 41, 42, 53]. Both weight-bearing endurance exercise and
resistance exercise have been found to generate increases in BMD. Of the 15
meta-analyses that evaluated the effects of exercise training on lumbar spine
BMD, 11 reported significant benefits (see Table 1). In general, the weighted
mean differences between exercisers and controls indicate that exercise training
can improve lumbar spine BMD by 1–2%. The meta-analyses seemingly further
suggest that exercise training is more likely to improve spine BMD than hip
BMD. Only eight of 15 meta-analyses reported significant benefits of exercise
training on BMD of the femoral neck or some other region of the proximal
femur (see Table 1). The greater responsiveness of lumbar spine BMD to
exercise, as compared with the proximal femur, could be related to the high
proportion of cancellous bone in the spine, which has a higher turnover rate than
cortical bone. Alternatively, because the proximal femur is loaded during all
physical activities performed in a standing posture, it is possible that exercise
training interventions do not generate strains in the proximal femur that are
markedly different from peak strains that occur during usual daily activities. This
may be particularly true for resistance exercise interventions if the majority of
exercises are performed in a seated position.
2.1 Exercise and Bone Strength
BMD measured by dual-energy X-ray absorptiometry (DXA) is the best single

predictor of fracture risk, but it accounts for only about 60% of the variability in
bone strength [94]. Other characteristics of bone, such as the shape, distribution of
mass, and trabecular microarchitecture, are also determinants of the resistance to
Table 1 Meta-analyses of the effects of exercise training on spine and hip bone mineral density (BMD, g/cm2)
240
Reference Subjects Primary outcomes Studies included Major results: effect size (95% CI)
[8] Postmenopausal women, LS, hip BMD 5 RCTs LS BMD 0.34 (-0.19–0.88)
50+ years, without 13 CTs FN BMD 0.35 (-0.47–1.17)
osteoporosis Troch BMD 0.85 (-0.10–1.80)
[9] Postmenopausal women LS, hip BMD 18 RCTs LS BMD 1.79 (0.58–3.01)
Hip BMD 0.68 (-1.18–2.53)
[38] Postmenopausal women LS BMD 4 RCTs LS BMD 2.83 (1.33–4.35)
6 CTs (ET studies)
[39] Postmenopausal women Hip BMD 2 RCTs Hip BMD 0.43 (0.04–0.81)
4 CTs (ET studies) Equivalent to a between-group difference in BMD of 2.4%
(2.1% in exercisers, -0.3% in controls)
[40] Postmenopausal women Regional BMD 11 RCTs (ET, RT studies) Overall 0.27 (0.16–0.37)
ET 1.62 (1.12–2.12)
RT 0.65 (0.48–0.83)
[43] Men BMD, any region 2 RCTs Overall 0.028 (-0.166–0.230)
6 CTs LS BMD 0.749 (0.099–1.327)
Femur 0.482 (0.270–0.705)
[44] Pre- and postmenopausal LS, femur, BMD 18 RCTs (RT studies) LS BMD 0.24 (0.11–0.38)
women Femur BMD 0.07 (-0.20–0.15)
Equivalent to 1.3% for the LS and 0.4% for the femur
[45] Postmenopausal women LS BMD 7 RCTs LS BMD:
6 CTs Exercisers 0.005 ± 0.043
Controls -0.007 ± 0.045
Interaction effect, P \ 0.001
Group effect, P = 0.003
Equivalent to a 2% benefit in LS BMD (1% in exercisers,
-1% in controls)
(continued)
W. M. Kohrt et al.
Table 1 (continued)
[41] Premenopausal women LS, FN BMD 3 CTs (individual data) LS BMD (g/cm2):
Exercisers 0.006 ± 0.035
Controls 0.008 ± 0.091
FN BMD (g/cm2):
Exercisers 0.005 ± 0.031
Controls 0.003 ± 0.031
No significant effects
[42] Postmenopausal women FN BMD 5 RCTs FN BMD (g/cm2):
5 CTs (individual data) Exercisers 0.004 ± 0.039
Controls 0.001 ± 0.048
No significant effects
[53] Postmenopausal women and elderly men Any BMD 5 RCTs No significant effects
2 CTs (tai chi studies)
[61] Postmenopausal women LS, FN, total hip BMD 15 RCTs (RT studies) LS BMD 0.006 (0.002–0.011)
FN BMD 0.010 (-0.002–0.021)
Hip BMD 0.002 (-0.001–0.005)
[62] Premenopausal women LS, FN BMD 6 RCTs, 1 CT (RT studies) LS BMD 0.014 (0.009–0.019)
Effects of Exercise and Physical Interventions on Bone
FN BMD 0.001 (-0.006–0.008)

[58] Postmenopausal women LS, FN BMD 5 RCTs, 3 CTs (walking studies) LS BMD 0.007 (-0.001–0.016)
FN BMD 0.014 (0.000–0.028)
[59] Postmenopausal women LS, FN, total hip BMD 10 RCTs LS BMD 0.015 (0.005–0.025)
5 CT (impact exercise) FN BMD 0.008 (0.004–0.013)
Hip BMD 0.013 (0.001–0.024)
[60] Premenopausal women LS, FN BMD 6 RCTs, 3 CTs (impact exercise) LS BMD 0.006 (0.002–0.010)
FN BMD 0.012 (0.005–0.020)
(continued)
241
Table 1 (continued)
242
[93] Pre- and postmenopausal women LS, FN BMD 32 RCTs (ET, RT studies) Annualized changes (%/year)
Premenopausal:
LS BMD
ET 1.5 (0.6–2.4)
RT 1.3 (0.8–1.8)
FN BMD
ET 0.7 (-0.3–1.7)
RT Inadequate data
Postmenopausal:
LS BMD
ET 1.3 (0.7–1.9)
RT 1.0 (0.4–1.6)
FN BMD
ET 0.5 (0.1–0.9)
RT 1.4 (0.2–2.6)
[95] Pre- and postmenopausal women LS, FN BMD 16 RCTs Annualized changes (%/year)
9 CTs (ET, RT studies) RCTs, premenopausal:
LS BMD
ET ? RT 0.9 (0.4–1.4)
FN BMD
ET ? RT 0.9 (0.3–1.5)
CTs, premenopausal:
LS BMD
ET ? RT 0.9 (-0.3–2.1)
FN BMD
ET ? RT Inadequate data
RCTs, postmenopausal:
LS BMD
ET 1.0 (0.4–1.5)
RT 0.4 (-0.3–1.2)
W. M. Kohrt et al.
(continued)
Table 1 (continued)
ET ? RT 0.8 (0.3–1.2)
FN BMD
ET 0.9 (0.3–1.5)
RT 0.9 (-0.2–1.9)
ET ? RT 0.9 (0.4–1.4)
CTs, Postmenopausal:
LS BMD
ET 2.2 (1.8–2.7)
RT Inadequate data
ET ? RT 2.4 (2.0–2.8)
FN BMD
ET 1.9 (0.8–2.9)
RT Inadequate data
ET ? RT 1.7 (0.6–2.7)
Results are given as normalized mean effect size, with 95% confidence interval in parentheses; an effect size of 0 indicates no effect of exercise; if the 95% confidence interval is
greater than 0, a positive effect of exercise is supported
BMD bone mineral density, CT controlled trial, ET endurance training, FN femoral neck, LS lumbar spine, RCT randomized controlled trial, RT resistance training, Troch
Effects of Exercise and Physical Interventions on Bone
trochanter
243
fracture. Bone strength cannot be directly measured in humans, but imaging

techniques (e.g., peripheral quantitative computed tomography (pQCT), magnetic
resonance imaging (MRI), DXA hip structural analysis, QCT-based finite element
analysis) can generate surrogates of strength. A meta-analysis of the few studies
that evaluated the effects of exercise training on estimates of bone strength
concluded that exercise does not significantly increase bone strength in adults [72].
However, this should be interpreted cautiously because the studies included in
the meta-analysis did not necessarily find the expected benefits of exercise training
on BMD.
On average, only small increases in BMD of 1–2% can be generated in adults
through exercise training. This magnitude of change is similar to or smaller than the
changes in BMD that occur in response to pharmacologic therapies that have
proven anti-fracture efficacy, including estrogens [33], raloxifene [19], bisphos-
phonates [35], denosumab [18], and parathyroid hormone [69]. In this context and
in the absence of strong evidence from RCTs for anti-fracture benefits of exercise, it
might be argued that exercise is unlikely to be as effective in preventing osteopo-
rotic fractures as pharmacologic therapy. However, there are two important con-
siderations for why exercise should continue to be avidly promoted for the
prevention and treatment of osteoporosis. First, exercise is effective in reducing risk
for falling [26] by improving such factors as balance and strength, which are not
targets of pharmacologic osteoporosis therapies. Second, it is possible that exercise
improves bone strength in a more structurally and functionally appropriate manner
than pharmacologic therapies. Evidence to support this comes from preclinical
studies of laboratory animals, which indicate that bisphosphonates [63] and para-
thyroid hormone [31] generate relative increases in bone strength that are roughly
proportional to the relative increases in bone mineral content (BMC) and BMD.
For example, bisphosphonate therapy increased BMC and BMD by *15% and
increased ultimate force and energy to failure by 15–20% [63]. In contrast,
mechanical loading of the rat forelimb resulted in smaller increases in ulnar whole-
bone BMC and BMD of *6%, but these were associated with disproportionately
large increases in ultimate force and energy to failure of 60–90% [81]. This
remarkable improvement in bone strength occurred because mechanical loading led
to a localized increase in bone area and moment of inertia at the site of peak strain.
Thus, mechanical loading can target specific bone regions whereas systemically
administered drugs presumably affect all regions of the skeleton, including non-
load-bearing regions. This suggests that exercise can selectively strengthen skeletal
regions that are susceptible to osteoporotic fracture. However, the extent to which
pre-clinical findings are of clinical relevance remains unclear. The level of strain
that generated the large increases in bone strength in rats was 2,000–3,000 le [81].
Based on in vivo measurements of tibial strain, it seems unlikely that strains of this
magnitude are regularly achieved by humans during physical activities [11, 66, 67].
There in an unmet need to devise exercise protocols that can safely stimulate bone
formation (or diminish bone loss) at sites that are prone to fracture.
Observational studies suggest that muscle power and physical activity are
directly, but modestly, associated with bone strength in older women and men
[1, 17]. For example, among 1,171 men aged 65+ years in the Osteoporotic
Fractures in Men (MrOS) study, the pQCT-derived bone strength index of the tibia
was 7% higher in the most physically active quartile than in the least active quartile,
and 5% higher in the quartile with the highest leg muscle power than in the lowest
quartile [17]. Based on this evidence, the magnitude of the benefit of physical
activity on bone strength in humans is likely considerably less than suggested by
the preclinical studies of mechanical loading [81]. The effects of physical activity
on fracture risk should reflect the net effects on bone strength and fall risk and can
be best evaluated through a randomized controlled intervention approach.
3 Key Determinants of the Adaptive Response

of Bone to Loading
3.1 Conventional Loading Factors
Preclinical studies of laboratory animals in the 1980s demonstrated that the

adaptive response of bone to mechanical loading is favorably influenced when
strains are dynamic rather than static, are of high peak magnitude and rate, and
present a unique strain distribution [91]. Further, when these conditions are met,
few loading repetitions are necessary to optimize the adaptive response. In general,
these principles also have been found to be of clinical relevance. Intervention trials
of exercise training indicate that activities that generate relatively high-intensity
bone-loading forces are more likely to result in an increase in BMD than those that
generate low-intensity forces [29, 48]. Similarly, cross-sectional comparisons of
adults who participate in different types of sports indicate that activities that
generate high-impact (e.g., volleyball) or odd-impact loading forces (e.g., soccer)
are associated with high BMD whereas activities that involve non-impact loading
(e.g., swimming) are not [29, 48, 73].
Mechanical stresses are introduced to the skeleton through a combination of
external forces (e.g., gravity, ground-reaction, and other external contacts) and
internal forces (e.g., muscle, joint-reaction) [6, 36, 46, 78]. Any physical activity
will result in some level of muscle force to achieve dynamic equilibrium (i.e.,
produce acceleration and/or provide joint stability). The magnitude of the muscle
forces can range from moderate, such as in endurance weight-bearing activities
(e.g., running), to high, such as in explosive jumping activities (e.g., volleyball)
or activities that involve quick accelerations or directional changes (e.g., soccer).
It has been suggested that muscle forces account for the majority of the adaptive
response of the skeleton to exercise [78]. Indeed, among 685 women and men aged
20–91 years, total body bone mineral content (BMC) was more strongly associated
with total body fat-free mass (i.e., FFM, a surrogate of muscle mass; r = 0.68 in
women and r = 0.63 in men) than with fat mass (women, r = 0.22; men, r = 0.14)
or total body mass (r = 0.48; men, r = 0.45) [46]. The strong association of BMC
with FFM seemingly provides compelling support for the notion that muscle forces
are the primary mediators of the effects of mechanical loading on the skeleton. Yet
activities that are effective in building muscle mass are not necessarily more
beneficial than non-muscle building activities. For example, intervention trials that
compared weight-bearing endurance and resistance exercise found that both modes
of training resulted in significant and similar increase in total hip and lumbar spine
BMD, despite the fact that only resistance training generated an increase in FFM
[49, 89]. Further, femoral neck BMD increased in response to weight-bearing
endurance exercise but not resistance exercise [49]. This suggests that resistance
training does not necessarily replicate the local skeletal stimulus produced by
weight-bearing activities. Additional studies are needed to clarify the aspects of
each training mode that are beneficial to the skeleton.
3.2 Additional Loading Factors
3.2.1 Novel Loading Factors
Preclinical research has revealed additional determinants of the skeletal adaptation

to loading that have not yet been extensively evaluated in clinical studies. One
important concept that has emerged is that bones appear to become desensitized to
mechanical stress after only a few loading cycles [80]. In rats, the mechanosensi-
tivity of bone is reduced by approximately 95% after only 20 loading cycles that
engender a high level of strain. However, interposing rest intervals of at least 4 h
between loading sessions can more than double the bone formation response, as
compared with applying the same number of loading cycles in a single session. This
suggests that multiple short bouts per day of bone-loading exercise would have more
favorable skeletal effects than a single longer bout. There is preliminary support for
this concept in humans. Jumping exercises performed twice per day over 8 weeks
resulted in a significant increase in a biomarker of bone formation whereas per-
forming the same number of jumps in a single session per day did not [20].
It has also been found in preclinical studies that the bone formation response to
loading can be augmented by interposing brief rest intervals of 10–15 s between
loading cycles [80, 90]. The clinical relevance of this for exercise prescription is
questionable. It would be possible to use this approach in an activity such as weight
lifting, but performing repetitions at 15 s intervals may have practical limitations.
However, there may be other approaches for building rest intervals into an exercise
session. For example, if a weight lifting prescription includes two sets of 10 dif-
ferent exercises, it might be more beneficial to do one set of each exercise and then
repeat the circuit, rather than doing two sets of an exercise before moving to the
next station. Rest intervals during a chronic loading intervention may also favorably
influence skeletal adaptations [85]. In rats, axial loading for 10 weeks out of a
15 week intervention (i.e., no loading during weeks 6–10) resulted in larger
increases in bone strength than axial loading for the entire 15 weeks.
Clinical studies to expand the understanding of how bones become desensitized

to loading will be critical for refining the exercise prescription for bone health.
However, such studies will be much more challenging to carry out in humans than
in laboratory animals because of the inability to measure the acute bone formation
response to loading. In this context, additional preclinical studies are needed to
facilitate the translation of preclinical to clinical research. One important question
to address is whether bone remains sensitive to loading when the loading condi-
tions are not predictable. Most experiments in this area have used extrinsic loading
models, with force applied at a consistent rate, intensity, and direction. Although
this approach provides desirable control of the mechanical environment [79], it is
of limited clinical relevance because such conditions do not occur during exercise
in humans. Another important question is whether there are biomarkers of the
acute desensitization of bone to loading that can be measured in humans. In the
absence of such biomarkers, clinical research in this area will progress slowly.
3.2.2 Aging
It has been suggested that the skeleton loses its responsiveness to mechanical loading
with advancing age [4, 24]. For example, 5 months of jumping exercises (50 vertical
jumps per day, 6 days/week) generated 2–3% increases in femoral neck and trochanter
BMD in premenopausal women [4], but the same intervention carried out for
11 months generated no increase in BMD in postmenopausal women not on meno-
pausal hormone therapy. The discordant responses were not explained by differences
in the peak ground-reaction forces during jumping, which tended to be higher in
postmenopausal women [4]. Although not significant, the BMD responses of post-
menopausal women to jumping were more favorable (1–2%) in women on estrogen-
based hormone therapy when compared with estrogen-deficient women. This suggests
that factors other than age, per se (e.g., reproductive status), influence the skeletal
response to mechanical loading. Preclinical research indicates that bone retains the
ability to adapt to mechanical loading into very old age (see ‘‘Skeletal mechanore-
sponsiveness and aging–animal studies’’), suggesting that there are skeletal benefits
of exercise across the life span. This is supported by clinical research demonstrating
that older women and men can increase BMD in response to exercise training [49, 52].
Thus, the aged skeleton appears to remain responsive to mechanical loading, but the
stimulus to generate an increase in BMD may change with aging. Whether the physical
activity recommendations for bone health should be age-specific remains unclear. It is
possible that physical activity has beneficial effects on bone strength and fall risk even
in the absence of measureable benefits on BMD.
3.3 Non-Loading Factors
Meta-analyses suggest that exercise training can generate average increases in BMD
of 1–3% in humans (Table 1). As a general observation, such changes seem small
when considered against the effectiveness of mechanical stress to activate bone

formation in preclinical research models. One reason for this may be that the strains
engendered by exercise in clinical studies are of lower magnitude than in preclinical
studies. Another possibility is that the difference between exercise-induced strains
and strains engendered during other daily activity is smaller in free-living humans
than in caged animals. Alternatively, there are a myriad of factors that could mediate
or moderate the adaptive skeletal response to loading, including, but not limited to,
hormones [e.g., parathyroid hormone (PTH)], minerals (e.g., calcium), growth fac-
tors (e.g., insulin-like growth factor), cytokines [e.g., interleukin 6 (IL-6)], and
signaling factors (e.g., prostaglandins). A few examples of factors that have potential
to influence the skeletal response to exercise are discussed below.
3.3.1 Nonsteroidal Anti-Inflammatory Drugs (NSAIDs)
Prostaglandin E2 (PGE2) increases in bone in response to mechanical loading

and is an essential signaling factor for bone formation [15, 96]. The key enzyme
for prostaglandin production is cyclooxygenase (COX), which is inhibited by
NSAIDs. There is consistent and compelling evidence that NSAIDs attenuate the
bone formation response to mechanical stress in laboratory animals [16, 23, 54].
A limitation of the preclinical studies is that they have used acute loading only.
However, if NSAIDs consistently block bone formation in response to repeated
bouts of loading, this would suggest that the benefits of exercise training on BMD
may not be fully realized in people who use NSAIDs. Observational studies of the
association of NSAID use with BMD or fracture risk have been mixed, demon-
strating adverse, neutral, or even beneficial effects of NSAID use [5, 13, 68, 76, 92].
However, because PGE2 can stimulate both bone formation and resorption, NSAIDs
may have both anti-formation and anti-resorptive effects on bone metabolism [75].
Importantly, the timing of NSAID administration is a key determinant of the bone
formation response, which is attenuated when NSAIDs are given before mechanical
loading, but not when they are given after [16, 25, 54]. Thus, the mixed associations
of NSAID use with BMD in observational studies may reflect the fact that none of the
studies considered the reason for or timing of NSAID use (e.g., before exercise to
reduce musculoskeletal discomforts).
There has been one proof-of-concept clinical study to determine whether using
NSAIDs before exercise inhibits the adaptation of BMD to exercise training [47].
Premenopausal women underwent a 9-month exercise training intervention and took
either ibuprofen 400 mg or placebo before and after every exercise session. The
randomized treatment assignments were: (1) ibuprofen before exercise, placebo
after, (2) placebo before and after exercise, and (3) placebo before exercise and
ibuprofen after. Consistent with preclinical research, the least favorable adaptations
in BMD tended to occur in the group that took ibuprofen before exercise sessions.
However, an unexpected finding was that use of ibuprofen after exercise sessions
augmented the increases in BMD when compared with placebo. It is possible that
NSAID use after exercise permits the activation of the bone formation, but
Fig. 3 Hypothetical effects Sweating during exercise

of the disruption of calcium
homeostasis during exercise
on bone resorption
Dermal calcium loss
Serum calcium
Stabilization of PTH
serum calcium
Bone resorption
attenuates exercise-induced increases in inflammatory cytokines, such as IL-6,

which could activate bone resorption [64]. Further research will be needed to con-
firm the findings of this preliminary study and evaluate underlying mechanisms by
which NSAID use influences skeletal adaptations to exercise. Such research is of
high clinical significance because of the important role of exercise in the prevention
and treatment of osteoporosis and the widespread use of NSAIDs.
3.3.2 Disruption of Calcium Homeostasis During Exercise
Individuals who participate in weight-supported activities, such as cycling and

swimming, have often been observed to have BMD levels that are lower than those
of athletes who participate in weight-bearing activities [29]. This would not be
unexpected if, in general, weight-supported activities have less osteogenic
potential than weight-bearing activities. However, BMD levels of cyclists and
swimmers have sometimes been reported to be low even when compared with
normally active control subjects [12, 14, 70]. This suggests that the exercise may
be the cause of low BMD in some athletes. Indeed, a prospective study of
competitive road cyclists found that BMD declined over a year of training and
competition [3]. Total hip BMD decreased by 1.5%, a rate comparable to post-
menopausal bone loss, and was not explained by low calcium intake or changes in
body mass or composition. A prospective study of master cyclists also reported
that the rate of decline in BMD was greater in cyclists than in controls [71].
It is possible that disruptions in calcium homeostasis during exercise trigger
metabolic responses that lead to the loss of bone mineral. A working model for
this hypothesis (Fig. 3) portends that dermal calcium loss during exercise triggers
a decline in serum ionized calcium. Because serum calcium levels are well-
defended, the appropriate counter-regulatory response to a decline is an increase
in PTH, which can stabilize serum calcium levels by mobilizing skeletal calcium
(i.e., via stimulation of bone resorption), enhancing intestinal calcium absorption,

and decreasing urinary calcium excretion. There is growing evidence to support
several steps of this conceptual model. Dermal calcium loss clearly increases
during prolonged and or vigorous exercise, and may be more than 100 mg/h
[2, 3, 57]. Exercise has been found to trigger a decline in serum calcium
[2, 55, 56] and an increase in PTH [2, 3, 30, 86]. The decrease in serum calcium is
not a consistent finding, but may not be apparent at the end of exercise if counter-
regulatory mechanisms to prevent a decline in serum calcium are effective.
Finally, there is evidence that exercise can result in an acute increase in markers
of bone resorption [2, 30, 86]. Perhaps the most compelling evidence for the
biological plausibility of this cascade of events comes from a study in which
exercise was performed with or without calcium supplementation (i.e., calcium-
enriched water) before and during exercise [30]. In the absence of calcium sup-
plementation, exercise resulted in a 2- to 3-fold increase in serum PTH and a 48%
increase in serum C-terminal cross-linking telopeptide of type I collagen (CTX), a
marker of bone resorption. Calcium supplementation prevented the increases in
both PTH and CTX.
If the disruption in calcium homeostasis during exercise described in Fig. 3
occurs repeatedly during exercise training, this may contribute to the decline in
BMD that has been observed in cyclists during training and competition [3].
Further studies will be necessary to determine whether the decline in BMD can be
prevented by consuming calcium before and during exercise bouts. Although most
of the research on the disruption of calcium homeostasis during exercise has been
conducted on cyclists, it seems likely that it occurs during any exercise that is
sufficiently intense or prolonged to result in substantial dermal calcium loss.
It seems plausible that the net effect on BMD will depend on the relative activation
of both bone resorption (e.g., through an increase in PTH) and bone formation
during exercise. Under such a paradigm, the deleterious effects of disruptions in
calcium homeostasis would be more likely to adversely affect BMD in individuals
who participate in activities with a relatively low osteogenic potential (e.g.,
swimming) than in those with a relatively high osteogenic potential (e.g., soccer).
It is also important to recognize that, if dermal calcium loss during exercise is the
trigger for the metabolic cascade described in Fig. 3, preclinical studies of rodents
will be of little use in advancing the understanding of these mechanisms.
4 Summary
The available evidence from prospective cohort and case–control studies suggests
that 3–4 h/week of physical activity can reduce the risk of hip fracture by 30–40%.
Activities such as walking, which are generally not adequate to stimulate an
increase in BMD when evaluated in intervention trials, may provide fracture
protection by preserving BMD and bone strength and/or by reducing risk of falls.
Follow-up evaluations from two small RCTs support the notion that exercise
intervention reduces the incidence of fracture, but larger RCTs will be necessary to
confirm the anti-fracture efficacy of exercise.
Recommendations for the type of exercise likely to preserve bone health [48]
have evolved from preclinical investigations of the key factors that influence the
bone formation response to mechanical loading. Further optimization of the
exercise prescription for bone health will likely require mechanistically-driven
clinical studies that are guided by preclinical research. The translation of pre-
clinical findings to clinical investigation can be accelerated if preclinical studies
are focused not only on mechanistic underpinnings, but also on the potential
clinical implications of the findings.
In general, the public health message that exercise is beneficial for both the
prevention and treatment of osteoporosis has a firm foundation of supporting
evidence. However, it must also be recognized that, under certain conditions,
exercise may have unfavorable effects on bone metabolism. Therefore, it is
important that future clinical research efforts be focused not only on identifying
factors that optimize the skeletal adaptations to exercise, but also on factors that
may compromise the adaptive response.
References
1. Ashe, M.C., Liu-Ambrose, T.Y., Cooper, D.M., Khan, K.M., McKay, H.A.: Muscle power is
related to tibial bone strength in older women. Osteoporos. Int. 19(12), 1725–1732 (2008)
2. Barry, D.W., Hansen, K.C., Van Pelt, R.E., Witten, M., Wolfe, P., Kohrt, W.M.: Acute
calcium ingestion attenuates exercise-induced disruption of calcium homeostasis. Med. Sci.
Sports Exerc. (2010)
3. Barry, D.W., Kohrt, W.M.: BMD decreases over the course of a year in competitive male
cyclists. J. Bone Miner. Res. 23(4), 484–491 (2008)
4. Bassey, E.J., Rothwell, M.C., Littlewood, J.J., Pye, D.W.: Pre- and postmenopausal women
have different bone mineral density responses to the same high-impact exercise. J. Bone
Miner. Res. 13(12), 1805–1813 (1998)
5. Bauer, D.C., Orwoll, E.S., Fox, K.M., et al.: Aspirin and NSAID use in older women: effect
on bone mineral density and fracture risk. Study of Osteoporotic Fractures Research Group.
J. Bone Miner. Res. 11(1), 29–35 (1996)
6. Beck, B.R.: Muscle forces or gravity—what predominates mechanical loading on bone?
Introduction. Med. Sci. Sports Exerc. 41(11), 2033–2036 (2009)
7. Benetou, V., Orfanos, P., Benetos, I.S., et al.: Anthropometry, physical activity and hip
fractures in the elderly. Injury (2011, in press)
8. Berard, A., Bravo, G., Gauthier, P.: Meta-analysis of the effectiveness of physical activity for
the prevention of bone loss in postmenopausal women. Osteoporos. Int. 7(4), 331–337 (1997)
9. Bonaiuti, D., Shea, B., Iovine, R., et al.: Exercise for preventing and treating osteoporosis in
postmenopausal women. Cochrane Database Syst. Rev. 3, CD000333 (2002)
10. Boonyaratavej, N., Suriyawongpaisal, P., Takkinsatien, A., Wanvarie, S., Rajatanavin, R.,
Apiyasawat, P.: Physical activity and risk factors for hip fractures in Thai women.
Osteoporos. Int. 12(3), 244–248 (2001)
11. Burr, D.B., Milgrom, C., Fyhrie, D., et al.: In vivo measurement of human tibial strains
during vigorous activity. Bone 18(5), 405–410 (1996)
12. Campion, F., Nevill, A.M., Karlsson, M.K., et al.: Bone status in professional cyclists. Int.
J. Sports Med. 31(7), 511–515 (2010)
13. Carbone, L.D., Tylavsky, F.A., Cauley, J.A., et al.: Association between bone mineral density
and the use of nonsteroidal anti-inflammatory drugs and aspirin: impact of cyclooxygenase
selectivity. J. Bone Miner. Res. 18(10), 1795–1802 (2003)
14. Carbuhn, A.F., Fernandez, T.E., Bragg, A.F., Green, J.S., Crouse, S.F.: Sport and training
influence bone and body composition in women collegiate athletes. J. Strength Cond. Res.
24(7), 1710–1717 (2010)
15. Chow, J.W.: Role of nitric oxide and prostaglandins in the bone formation response to
mechanical loading. Exerc. Sport Sci. Rev. 28(4), 185–188 (2000)
16. Chow, J.W., Chambers, T.J.: Indomethacin has distinct early and late actions on bone
formation induced by mechanical stimulation. Am. J. Physiol. 267(2 Pt 1), E287–E292
(1994)
17. Cousins, J.M., Petit, M.A., Paudel, M.L., et al.: Muscle power and physical activity are
associated with bone strength in older men: the osteoporotic fractures in men study. Bone
47(2), 205–211 (2010)
18. Cummings, S.R., San, M.J., McClung, M.R., et al.: Denosumab for prevention of fractures in
postmenopausal women with osteoporosis. N. Engl. J. Med. 361(8), 756–765 (2009)
19. Ensrud, K.E., Stock, J.L., Barrett-Connor, E., et al.: Effects of raloxifene on fracture risk in
postmenopausal women: the Raloxifene use for the heart trial. J. Bone Miner. Res. 23(1),
112–120 (2008)
20. Erickson, C.R., Vukovich, M.D.: Osteogenic index and changes in bone markers during a
jump training program: a pilot study. Med. Sci. Sports Exerc. 42(8), 1485–1492 (2010)
21. Farahmand, B.Y., Persson, P.G., Michaelsson, K., et al.: Physical activity and hip fracture: a
population-based case-control study. Int. J. Epidemiol. 29(2), 308–314 (2000)
22. Feskanich, D., Willett, W., Colditz, G.: Walking and leisure-time activity and risk of hip
fracture in postmenopausal women. J. Am. Med. Assoc. 288(18), 2300–2306 (2002)
23. Forwood, M.R.: Inducible cyclo-oxygenase (COX-2) mediates the induction of bone
formation by mechanical loading in vivo. J. Bone Miner. Res. 11(11), 168–1693 (1996)
24. Forwood, M.R., Burr, D.B.: Physical activity and bone mass: exercises in futility? Bone
Miner 21(2), 89–112 (1993)
25. Forwood, M.R., Owan, I., Takano, Y., Turner, C.H.: Increased bone formation in rat tibiae
after a single short period of dynamic loading in vivo. Am. J. Physiol. 270(3 Pt 1), E419–
E423 (1996)
26. Gillespie, L.D., Robertson, M.C., Gillespie, W.J., et al.: Interventions for preventing falls in
older people living in the community. Cochrane Database Syst. Rev. 2, CD007146 (2009)
27. Gregg, E.W., Cauley, J.A., Seeley, D.G., Ensrud, K.E., Bauer, D.C.: Physical activity and
osteoporotic fracture risk in older women. Study of Osteoporotic Fractures Research Group.
Ann. Intern. Med. 129(2), 81–88 (1998)
28. Gregson, C.L., Carson, C., Amuzu, A., Ebrahim, S.: The association between graded physical
activity in postmenopausal British women, and the prevalence and incidence of hip and wrist
fractures. Age Ageing 39(5), 565–574 (2010)
29. Guadalupe-Grau, A., Fuentes, T., Guerra, B., Calbet, J.A.: Exercise and bone mass in adults.
Sports Med. 39(6), 439–468 (2009)
30. Guillemant, J., Accarie, C., Peres, G., Guillemant, S.: Acute effects of an oral calcium load on
markers of bone metabolism during endurance cycling exercise in male athletes. Calcif.
Tissue Int. 74(5), 407–414 (2004)
31. Hirano, T., Burr, D.B., Turner, C.H., Sato, M., Cain, R.L., Hock, J.M.: Anabolic effects of
human biosynthetic parathyroid hormone fragment (1–34), LY333334, on remodeling and
mechanical properties of cortical bone in rabbits. J. Bone Miner. Res. 14(4), 536–545
(1999)
32. Hoidrup, S., Sorensen, T.I., Stroger, U., Lauritzen, J.B., Schroll, M., Gronbaek, M.: Leisure-
time physical activity levels and changes in relation to risk of hip fracture in men and women.
Am. J. Epidemiol. 154(1), 60–68 (2001)
33. Jackson, R.D., Wactawski-Wende, J., LaCroix, A.Z., et al.: Effects of conjugated equine
estrogen on risk of fractures and BMD in postmenopausal women with hysterectomy: results
from the women’s health initiative randomized trial. J. Bone Miner. Res. 21(6), 817–828
(2006)
34. Jaglal, S.B., Kreiger, N., Darlington, G.A.: Lifetime occupational physical activity and risk of
hip fracture in women. Ann. Epidemiol. 5(4), 321–324 (1995)
35. Jansen, J.P., Bergman, G.J., Huels, J., Olson, M.: The efficacy of Bisphosphonates in the
prevention of vertebral, hip, and nonvertebral-nonhip fractures in osteoporosis: a network
meta-analysis. Semin. Arthr. Rheum. 40(4), 275–284 (2011)
36. Judex, S., Carlson, K.J.: Is bone’s response to mechanical signals dominated by gravitational
loading? Med. Sci. Sports Exerc. 41(11), 2037–2043 (2009)
37. Kanis, J., Johnell, O., Gullberg, B., et al.: Risk factors for hip fracture in men from southern
Europe: the MEDOS study. Osteoporos. Int. 9(1), 45–54 (1999)
38. Kelley, G.: Aerobic exercise and lumbar spine bone mineral density in postmenopausal
women: a meta-analysis. J. Am. Geriatr. Soc. 46(2), 143–152 (1998)
39. Kelley, G.A.: Aerobic exercise and bone density at the hip in postmenopausal women: a
meta-analysis. Prev. Med. 27(6), 798–807 (1998)
40. Kelley, G.A.: Exercise and regional bone mineral density in postmenopausal women: a meta-
analytic review of randomized trials. Am. J. Phys. Med. Rehabil. 77(1), 76–87 (1998)
41. Kelley, G.A., Kelley, K.S.: Efficacy of resistance exercise on lumbar spine and femoral neck
bone mineral density in premenopausal women: a meta-analysis of individual patient data.
J. Womens’ Health 13(3), 293–300 (2004)
42. Kelley, G.A., Kelley, K.S.: Exercise and bone mineral density at the femoral neck in
postmenopausal women: a meta-analysis of controlled clinical trials with individual patient
data. Am. J. Obstet. Gynecol. 194(3), 760–767 (2006)
43. Kelley, G.A., Kelley, K.S., Tran, Z.V.: Exercise and bone mineral density in men: a meta-
analysis. J. Appl. Physiol. 88(5), 1730–1736 (2000)
44. Kelley, G.A., Kelley, K.S., Tran, Z.V.: Resistance training and bone mineral density in
women: a meta-analysis of controlled trials. Am. J. Phys. Med. Rehabil. 80(1), 65–77 (2001)
45. Kelley, G.A., Kelley, K.S., Tran, Z.V.: Exercise and lumbar spine bone mineral density in
postmenopausal women: a meta-analysis of individual patient data. J. Gerontol. A Biol. Sci.
Med. Sci. 57(9), M599–M604 (2002)
46. Kohrt, W.M., Barry, D.W., Schwartz, R.S.: Muscle forces or gravity: what predominates
mechanical loading on bone? Med. Sci. Sports Exerc. 41(11), 2050–2055 (2009)
47. Kohrt, W.M., Barry, D.W., Van Pelt, R.E., Jankowski, C.M., Wolfe, P., Schwartz, R.S.:
Timing of ibuprofen use and bone mineral density adaptations to exercise training. J. Bone
Miner. Res. 25(6), 1415–1422 (2010)
48. Kohrt, W.M., Bloomfield, S.A., Little, K.D., Nelson, M.E., Yingling, V.R.: American
College of Sports Medicine Position Stand: physical activity and bone health. Med. Sci.
Sports Exerc. 36(11), 1985–1996 (2004)
49. Kohrt, W.M., Ehsani, A.A., Birge Jr., S.J.: Effects of exercise involving predominantly either
joint-reaction or ground-reaction forces on bone mineral density in older women. J. Bone
Miner. Res. 12(8), 1253–1261 (1997)
50. Korpelainen, R., Keinanen-Kiukaanniemi, S., Nieminen, P., Heikkinen, J., Vaananen, K.,
Korpelainen, J.: Long-term outcomes of exercise: follow-up of a randomized trial in older
women with osteopenia. Arch. Intern. Med. 170(17), 1548–1556 (2010)
51. Kujala, U.M., Kaprio, J., Kannus, P., Sarna, S., Koskenvuo, M.: Physical activity and
osteoporotic hip fracture risk in men. Arch. Intern. Med. 160(5), 705–708 (2000)
52. Kukuljan, S., Nowson, C.A., Sanders, K.M., et al.: Independent and combined effects of
Calcium-Vitamin D3 and exercise on Bone structure and strength in older Men: an 18-month
factorial design randomized controlled trial. J. Clin. Endocrinol. Metab. 96(4), 955–963
(2011)
53. Lee, M.S., Pittler, M.H., Shin, B.C., Ernst, E.: Tai chi for osteoporosis: a systematic review.
Osteoporos. Int. 19(2), 139–146 (2008)
54. Li, J., Burr, D.B., Turner, C.H.: Suppression of prostaglandin synthesis with NS-398 has
different effects on endocortical and periosteal bone formation induced by mechanical
loading. Calcif. Tissue Int. 70(4), 320–329 (2002)
55. Maimoun, L., Simar, D., Caillaud, C., et al.: Response of calciotropic hormones and bone
turnover to brisk walking according to age and fitness level. J. Sci. Med. Sport 12(4), 463–
467 (2009)
56. Maimoun, L., Simar, D., Malatesta, D., et al.: Response of bone metabolism related hormones
to a single session of strenuous exercise in active elderly subjects. Br. J. Sports Med. 39(8),
497–502 (2005)
57. Martin, B.R., Davis, S., Campbell, W.W., Weaver, C.M.: Exercise and calcium
supplementation: effects on calcium homeostasis in sportswomen. Med. Sci. Sports Exerc.
39(9), 1481–1486 (2007)
58. Martyn-St James, M., Carroll, S.: A meta-analysis of impact exercise on postmenopausal bone
loss: the case for mixed loading exercise programmes. Br. J. Sports Med. (2008, in press)
59. Martyn-St James, M., Carroll, S.: A meta-analysis of impact exercise on postmenopausal
bone loss: the case for mixed loading exercise programmes. Br. J. Sports Med. 43(12), 898–
908 (2009)
60. Martyn-St James, M., Carroll, S.: Effects of different impact exercise modalities on bone
mineral density in premenopausal women: a meta-analysis. J. Bone Miner. Metab. 28(3),
251–267 (2010)
61. Martyn-St, J.M., Carroll, S.: High-intensity resistance training and postmenopausal bone loss:
a meta-analysis. Osteoporos. Int. 17(8), 1225–1240 (2006)
62. Martyn-St, J.M., Carroll, S.: Progressive high-intensity resistance training and bone mineral
density changes among premenopausal women: evidence of discordant site-specific skeletal
effects. Sports Med. 36(8), 683–704 (2006)
63. Mashiba, T., Turner, C.H., Hirano, T., Forwood, M.R., Johnston, C.C., Burr, D.B.: Effects of
suppressed bone turnover by bisphosphonates on microdamage accumulation and
biomechanical properties in clinically relevant skeletal sites in beagles. Bone 28(5), 524–
531 (2001)
64. McLean, R.R.: Proinflammatory cytokines and osteoporosis. Curr. Osteoporos. Rep. 7(4),
134–139 (2009)
65. Michaelsson, K., Olofsson, H., Jensevik, K., et al.: Leisure physical activity and the risk of
fracture in men. PLoS Med. 4(6), 1094–1100 (2007)
66. Milgrom, C., Finestone, A., Levi, Y., et al.: Do high impact exercises produce higher tibial
strains than running? Br. J. Sports Med. 34(3), 195–199 (2000)
67. Milgrom, C., Finestone, A., Simkin, A., et al.: In vivo strain measurements to evaluate the
strengthening potential of exercises on the tibial bone. J. Bone Joint Surg. Br. 82(4), 591–594
(2000)
68. Morton, D.J., Barrett-Connor, E., Schneider, D.L.: Nonsteroidal anti-inflammatory drugs and
bone mineral density in older women: the Rancho Bernardo study. J. Bone Miner. Res.
13(12), 1924–1931 (1998)
69. Neer, R.M., Arnaud, C.D., Zanchetta, J.R., et al.: Effect of parathyroid hormone (1–34) on
fractures and bone mineral density in postmenopausal women with osteoporosis. N. Engl.
J. Med. 344(19), 1434–1441 (2001)
70. Nichols, J.F., Palmer, J.E., Levy, S.S.: Low bone mineral density in highly trained male
master cyclists. Osteoporos. Int. 14(8), 644–649 (2003)
71. Nichols, J.F., Rauh, M.J.: Longitudinal changes in bone mineral density in male master
cyclists and nonathletes. J. Strength Cond. Res. 25(3), 727–734 (2011)
72. Nikander, R., Sievanen, H., Heinonen, A., Daly, R.M., Uusi-Rasi, K., Kannus, P.: Targeted
exercise against osteoporosis: a systematic review and meta-analysis for optimising bone
strength throughout life. BMC Med. 8, 47 (2010)
73. Nikander, R., Sievanen, H., Heinonen, A., Kannus, P.: Femoral neck structure in adult female
athletes subjected to different loading modalities. J. Bone Miner. Res. 20(3), 520–528 (2005)
74. Physical Activity Guidelines Advisory Committee: Physical activity guidelines advisory
committee report. Washington, DC, U.S. Department of Health and Human Services (2008)
75. Raisz, L.G.: Potential impact of selective cyclooxygenase-2 inhibitors on bone metabolism in
health and disease. Am. J. Med. 110 (Suppl 3A), 43S–45S (2001)
76. Richards, J.B., Joseph, L., Schwartzman, K., Kreiger, N., Tenenhouse, A., Goltzman, D.: The
effect of cyclooxygenase-2 inhibitors on bone mineral density: results from the Canadian
Multicentre Osteoporosis Study. Osteoporos. Int. 17(9), 1410–1419 (2006)
77. Robbins, J., Aragaki, A.K., Kooperberg, C., et al.: Factors associated with 5-year risk of hip
fracture in postmenopausal women. J. Am. Med. Assoc. 298(20), 2389–2398 (2007)
78. Robling, A.G.: Is bone’s response to mechanical signals dominated by muscle forces? Med.
Sci. Sports Exerc. 41(11), 2044–2049 (2009)
79. Robling, A.G., Burr, D.B., Turner, C.H.: Skeletal loading in animals. J. Musculoskelet.
Neuronal Interact. 1(3), 249–262 (2001)
80. Robling, A.G., Castillo, A.B., Turner, C.H.: Biomechanical and molecular regulation of bone
81. Robling, A.G., Hinant, F.M., Burr, D.B., Turner, C.H.: Improved bone structure and strength
after long-term mechanical loading is greatest if loading is separated into short bouts. J. Bone
Miner. Res. 17(8), 1545–1554 (2002)
82. Roy, D.K., O’Neill, T.W., Finn, J.D., et al.: Determinants of incident vertebral fracture in
men and women: results from the European Prospective Osteoporosis Study (EPOS).
Osteoporos. Int. 14(1), 19–26 (2003)
83. Rubin, C.T., Lanyon, L.E.: Regulation of bone mass by mechanical strain magnitude. Calcif.
Tissue Int. 37(4), 411–417 (1985)
84. Samelson, E.J., Hannan, M.T., Zhang, Y., Genant, H.K., Felson, D.T., Kiel, D.P.: Incidence
and risk factors for vertebral fracture in women and men: 25-year follow-up results from the
population-based Framingham study. J. Bone Miner. Res. 21(8), 1207–1214 (2006)
85. Saxon, L.K., Robling, A.G., Alam, I., Turner, C.H.: Mechanosensitivity of the rat skeleton
decreases after a long period of loading, but is improved with time off. Bone 36(3), 454–464
(2005)
86. Scott, J.P., Sale, C., Greeves, J.P., Casey, A., Dutton, J., Fraser, W.D.: The role of exercise
intensity in the bone metabolic response to an acute bout of weight-bearing exercise. J. Appl.
Physiol. 110(2), 423–432 (2011)
87. Silman, A.J., O’Neill, T.W., Cooper, C., Kanis, J., Felsenberg, D.: Influence of physical
activity on vertebral deformity in men and women: results from the European Vertebral
Osteoporosis study. J. Bone Miner. Res. 12(5), 813–819 (1997)
88. Sinaki, M., Itoi, E., Wahner, H.W., et al.: Stronger back muscles reduce the incidence of
vertebral fractures: a prospective 10 year follow-up of postmenopausal women. Bone 30(6),
836–841 (2002)
89. Snow-Harter, C., Bouxsein, M.L., Lewis, B.T., Carter, D.R., Marcus, R.: Effects of resistance
and endurance exercise on bone mineral status of young women: a randomized exercise
intervention trial. J. Bone Miner. Res. 7(7), 761–769 (1992)
90. Srinivasan, S., Weimer, D.A., Agans, S.C., Bain, S.D., Gross, T.S.: Low-magnitude
mechanical loading becomes osteogenic when rest is inserted between each load cycle.
J. Bone Miner. Res. 17(9), 1613–1620 (2002)
91. Turner, C.H., Pavalko, F.M.: Mechanotransduction and functional response of the skeleton to
physical stress: the mechanisms and mechanics of bone adaptation. J. Orthop. Sci. 3(6), 346–
355 (1998)
92. van Staa, T.P., Leufkens, H.G., Cooper, C.: Use of nonsteroidal anti-inflammatory drugs and
risk of fractures. Bone 27(4), 563–568 (2000)
93. Wallace, B.A., Cumming, R.G.: Systematic review of randomized trials of the effect of
exercise on bone mass in pre- and postmenopausal women. Calcif. Tissue Int. 67(1), 10–18
(2000)
94. Weinstein, R.S.: True strength. J. Bone Miner. Res. 15(4), 621–625 (2000)
95. Wolff, I., van Croonenborg, J.J., Kemper, H.C., Kostense, P.J., Twisk, J.W.: The effect of
exercise training programs on bone mass: a meta-analysis of published controlled trials in
pre- and postmenopausal women. Osteoporos. Int 9(1), 1–12 (1999)
96. Zaman, G., Suswillo, R.F., Cheng, M.Z., Tavares, I.A., Lanyon, L.E.: Early responses to
dynamic strain change and prostaglandins in bone-derived cells in culture. J. Bone Miner.
Res. 12(5), 769–777 (1997)
Author Index
A M
Akkus, Ozan, 105 Melville, Katherine M., 217
Allen, Matthew R., 151
Anderson, Dennis E., 133
P
Parkinson, Ian H., 31
B
Barry, Daniel W., 235
Bouxsein, Mary L., 133 S
Burr, David B., 151 Silva, Matthew J., 1, 191
F V
Fazzalari, Nicola L., 31 van der Meulen, Marjolein C. H., 217
Vashishth, Deepak, 87
Villalon, Karen L., 235
G
Genetos, Damian C., 177
W
Wang, Xiaodu, 53
J
Jacobs, Christopher R., 177
Jepsen, Karl J., 1 Y
Yerramshetty, Janardhan, 105
K
Karim, Lamya, 87
Kelly, Natalie H., 217
Kohrt, Wendy M., 235
Kotiya, Akhilesh A., 191

DOI: 10.1007/978-3-642-18053-8

Scientific Article

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Scientific Article

Uploaded by

Copyright:

Available Formats

Studies in Mechanobiology, Tissue Engineering

For further volumes:

ISSN 1868-2006 ISSN 1868-2014 (electronic)

Library of Congress Control Number: 2012943364

Ó Springer-Verlag Berlin Heidelberg 2013

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Age-Related Changes in Whole-Bone Structure and Strength . . . . . . . 1

Characterisation of Trabecular Bone Structure . . . . . . . . . . . . . . . . . 31

Cortical Bone Mechanics and Composition:

Bone Microdamage and Its Contributions to Fracture . . . . . . . . . . . . 87

Changes in Cortical Bone Mineral and Microstructure

Factor of Risk for Fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Bisphosphonates and PTH for Preventing Fractures . . . . . . . . . . . . . . 151

Bone Cell Mechanoresponsiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

The Effect of Aging on Skeletal Mechanoresponsiveness:

Skeletal Mechanoresponsiveness: Effects of Sex Hormones . . . . . . . . . 217

Effects of Exercise and Physical Interventions on Bone:

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257

Matthew J. Silva and Karl J. Jepsen

Abstract We review data on age-related changes in bone geometry of relevance

Stud Mechanobiol Tissue Eng Biomater (2013) 5: 1–30 1

equivalent rates of decline. In conclusion, there are important age-related changes

2 Diaphyseal Changes with Age

2.1 Cross-Sectional Geometry

while resorption prevails at the endocortex. Generally, these changes occur

2.1.1 Lower Extremity: Femur and Tibia

2.1.2 Upper Extremity: Metacarpals, Radius and Ulna

2.2 Structural Mechanical Properties

2.3 Coordinated Variations in Geometric and Material

Functional adaptation within the skeletal system can coordinate morphological

The variation in bone size relative to body size is likely to be compensated,

3 Metaphyseal and Vertebral Changes with Age

3.1 Cross-Sectional Geometry and Bone Mineral Density

3.1.1 Lower Extremity: Femoral Neck and Distal Tibia

% Change per decade*

3.1.2 Upper Extremity: Distal Radius

The distal radius is of clinical relevance as a common site of osteoporotic frac-

% Change per decade*

3.2 Metaphyseal and Vertebral Bone Strength

% Change per decade*

of model predictions to mechanical properties in cadaveric samples, these methods

3.2.1 Lower Extremity: Femoral Neck and Distal Tibia

3.2.2 Upper Extremity: Distal Radius

Mosekilde and Mosekilde [55] performed compression tests on thoracic and

Ian H. Parkinson and Nicola L. Fazzalari

Abstract The characterisation of trabecular bone structure has until recently

I. H. Parkinson (&) N. L. Fazzalari

Stud Mechanobiol Tissue Eng Biomater (2013) 5: 31–51 31

menopausal or age-related bone loss for females and males, respectively, is a

Fig. 1 X-ray image of a

2 Skeletal Distribution of Trabecular Bone: What and Where

Histological studies have shown, as expected, that there is considerable heter-

Fig. 3 Composite photomicrograph showing histological sections of trabecular bone from

Availability of human trabecular bone samples has always and continues to be a

3 Histomorphometry of Trabecular Bone Structure

where customised or commercial (Bioquant, Bioquant Image Analysis Corpora-

4 Non-Destructive Imaging and Morphometry

Notwithstanding the conceptual models developed to extrapolate measurements

The availability of 3D voxel-based datasets of trabecular bone has driven the

5 Trabecular Bone Structure in the Young

Notwithstanding the genetic or environmental effects on trabecular bone

6 Trabecular Bone Structure in the Adult

7 Trabecular Bone Structure in Older Age