Bone 53 (2013) 451–458

Contents lists available at SciVerse ScienceDirect

Bone
journal homepage: www.elsevier.com/locate/bone

Original Full Length Article

Longitudinal elastic properties and porosity of cortical bone tissue vary

with age in human proximal femur
M.K.H. Malo a, b,⁎, D. Rohrbach c, H. Isaksson a, d, J. Töyräs a, b, J.S. Jurvelin a, I.S. Tamminen a, e, f,
H. Kröger e, f, K. Raum c
a
Department of Applied Physics, University of Eastern Finland, POB 1627, FI-70211, Kuopio, Finland
b
Department of Clinical Neurophysiology, Kuopio University Hospital, POB 1777, FI-70211, Kuopio, Finland
c
Julius Wolff Institute & Berlin-Brandenburg School for Regenerative Therapies, Charité – Universitätsmedizin Berlin, Augustenburger Platz 1, DE-13353, Berlin, Germany
d
Division of Solid Mechanics, Department of Orthopaedics, Lund University, POB 118, SE-22100, Lund, Sweden
e
Bone and Cartilage Research Unit, University of Eastern Finland, POB 1627, FI-70211, Kuopio, Finland
f
Department of Orthopaedics, Traumatology, and Hand Surgery, Kuopio University Hospital, POB 1777, FI-70211, Kuopio, Finland

a r t i c l e i n f o a b s t r a c t

Article history: Tissue level structural and mechanical properties are important determinants of bone strength. As an individual
Received 20 August 2012 ages, microstructural changes occur in bone, e.g., trabeculae and cortex become thinner and porosity increases.
Revised 8 January 2013 However, it is not known how the elastic properties of bone change during aging. Bone tissue may lose its elasticity
Accepted 10 January 2013
and become more brittle and prone to fractures as it ages. In the present study the age-dependent variation in the
Available online 17 January 2013
spatial distributions of microstructural and microelastic properties of the human femoral neck and shaft were eval-
Edited by: David Burr uated by using acoustic microscopy. Although these properties may not be directly measured in vivo, there is a
major interest to investigate their relationships with the linear elastic measurements obtained by diagnostic
Keywords: ultrasound at the most severe fracture sites, e.g., the femoral neck. However, before the validity of novel in vivo
Bone techniques can be established, it is essential to understand the age-dependent variation in tissue elastic properties
Cortex and porosity at different skeletal sites. A total of 42 transverse cross-sectional bone samples were obtained
Ultrasound from the femoral neck (Fn) and proximal femoral shaft (Ps) of 21 men (mean±SD age 47.1±17.8, range
Scanning acoustic microscopy 17–82 years). Samples were quantitatively imaged using a scanning acoustic microscope (SAM) equipped with
a 50 MHz ultrasound transducer. Distributions of the elastic coefficient (c33) of cortical (Ct) and trabecular (Tr)
tissues and microstructure of cortex (cortical thickness Ct.Th and porosity Ct.Po) were determined. Variations in
c33 were observed with respect to tissue type (c33Tr b c33Ct), location (c33(Ct.Ps)=37.7 GPa>c33(Ct.Fn)=
35.3 GPa>c33(Tr.Ps)=33.8 GPa>c33(Tr.Fn)=31.9 GPa), and cadaver age (R2 =0.28–0.46, pb 0.05). Regional var-
iations in porosity were found in the neck (superior 13.1%; inferior 6.1%; anterior 10.1%; posterior 8.6%) and in the
shaft (medial 9.5%; lateral 7.7%; anterior 8.6%; posterior 12.0%). In conclusion, significant variations in elastic coef-
ficients were detected between femoral neck and shaft as well as between the quadrants of the cross-sections of
neck and shaft. Moreover, an age-related increase in cortical porosity and a stiffening of the bone tissue were ob-
served. These findings may explain in part the increase in susceptibility to suffer low energy fractures during aging
and highlight the potential of ultrasound in clinical osteoporosis diagnostics.
© 2013 Elsevier Inc. All rights reserved.

Introduction
The risk of sustaining a fracture increases with age [1,2]. In particular,
fractures of the proximal femur increase morbidity, shorten life expec-
tancy, and represent an economic burden of the health care system
[3,4]. The prevalence of fractures depends on both internal (quality of
the musculoskeletal system) and external risk factors (e.g., decreased
balance, limited vision or slipperiness of the ground). The external fac-
tors are believed to have a minor or even no effect on the mechanical
⁎ Corresponding author at: Department of Applied Physics, University of Eastern Finland,
POB 1627, FI-70211, Kuopio, Finland. Fax: +358 17 162585.
E-mail address: markus.malo@uef.fi (M.K.H. Malo).
strength of bone tissue, but are related to the number of falls and thereby
increasing the number of fractures [5–7]. On the contrary, bone quality,
i.e., the sum of factors such as bone geometry, degree of mineralization,
microstructure and mechanical (elastic) properties, determine the
mechanical strength and fragility of bone [8–10]. As these important
properties cannot be tested in vivo, there is major interest in their estima-
tion from linear elastic measurements obtained by diagnostic ultrasound
at the most severe fracture sites, e.g., the femoral neck. Indeed, there is
excellent evidence that quantitative ultrasound is a valid (radiation-free
and inexpensive) method for fracture risk assessment [11]. However,
before one can establish the validity of these emerging technologies, it
is essential to understand the age-dependent variation in tissue elastic
properties and porosity at different skeletal sites. Since these properties

8756-3282/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.bone.2013.01.015
452 M.K.H. Malo et al. / Bone 53 (2013) 451–458

directly influence the tissue acoustic properties, e.g., as measured by Table 1

non-invasive ultrasound. Furthermore, assessment of bone properties Basic anthropometric data of the cadavers.

at the tissue level may improve the understanding of the alterations Age [years] Age group Height [cm] Weight [kg] BMI [kg/m2]
occurring in the tissue due to aging, medication, or pathology, and may
17 1 178 74 23.4
further help to clarify more accurately the contribution of each internal 22 1 186 106 30.6
factor toward the overall fracture risk. 26 1 175 60 19.6
It is possible to obtain highly localized direct information of tissue 30 1 184 105 31.0
32 1 171 69 23.6
elastic and viscoelastic properties by means of nanoindentation
34 1 187 102 29.2
[12,13]. However, the technique is not truly capable of assessing the 37 1 177 74 23.6
elastic properties at high resolution over a large area such as the 39 1 185 84 24.5
cross-section of the femoral neck. Quantitative acoustic microscopy 43 2 171 98 33.5
has been widely used for assessing the elastic properties of bone in 44 2 179 96 30.0
46 2 185 85 24.8
vitro [14–18]. By measuring bone samples over a wide range of frequen-
49 2 178 85 26.8
cies and focal spot sizes, one can obtain both micro- and macro-scale as- 51 2 185 108 31.6
sessments of bone [19,20]. The ultrasound reflection amplitude relates 52 2 180 136 42.0
to the average elastic properties of the investigated material at the 54 2 176 73 23.6
58 2 169 96 33.6
focal spot. Since structures smaller than the spot size cannot be
59 2 175 73 23.8
detected, the transducer must be chosen carefully, e.g. when assessing 63 3 170 68 23.5
porosity of cortical bone. In scanning acoustic microscopy (SAM) of 74 3 166 64 23.2
bone, focal spot diameters in the range of 1 to 150 μm have been ap- 78 3 177 72 23.0
plied [20,21]. In human cortical bone, a spatial resolution of 23 μm has 82 3 165 53 19.5
47.1 ± 17.8 – 177.1 ± 6.7 84.8 ± 20.0 26.9 ± 5.5
been proven to be sufficient to resolve Haversian canals and resorption
cavities [20]. Most commonly in SAM, the amplitude of the reflected ul-
trasound pulse is measured, and used as a measure of the acoustic re-
flectivity which is determined by the material density and elastic that freezing and thawing do not affect significantly the elastic proper-
properties in the probing direction [14,15]. The tissue elastic coefficient ties of bone tissue [29–31].
can be estimated from the acoustic impedance measurements [15]. For Transversal cross-sections were cut from the femoral neck (n=21)
these reasons, SAM has been used extensively in the quantitative eval- and shaft (n=21) with a band saw (KT-210, Koneteollisuus Oy, Helsinki,
uation of spatial variation in elastic properties and microarchitecture of Finland). A minor cut was made on the medial bone surface to enable
bone over the whole cross-sectional samples [22–24]. Furthermore, correct orientation of the sample during the measurements. Sections of
SAM has been used to obtain the material parameters which can be the shaft were cut perpendicular to the long axis of bone, below the
used in computational models of bone tissue mechanics [25]. trochanter minor. Sections from the middle of the neck were cut perpen-
Earlier studies using SAM have demonstrated the feasibility of dicular to the long axis of the neck (Fig. 1). Samples were dehydrated in
this technique in the assessment of both the microstructural and ethanol and embedded in polymethylmethacrylate (PMMA) according
the tissue elastic properties in fresh and embedded bone samples to standard protocols [14,19]. In order to obtain smooth planar sample
[16,18,20,21,24,26]. For example, variations of acoustic impedance surfaces, the samples were ground with a plane grinder (EXAKT 400CS,
and porosity have been analyzed in the human distal radius [27]. Exakt Apparatebau, Norderstedt, Germany) with successively decreasing
Furthermore, in previous studies the variation of bone acoustic im- grain size (ISO/FEPA grit: P500, P800, P1000, P1200 and P4000, Hermes
pedance, elastic properties and their directional dependencies Abrasives Ltd., Virginia Beach, VA and Struers A/S, Ballerup, Denmark).
(anisotropy) have been investigated in the femoral shaft of human The final polishing was performed with cloths containing 6 μm and
donors [14,17,18,26,28]. However, previously no systematic investi- 1 μm diamond particle suspensions in ethylene glycol (Phoenix 4000,
gation has been conducted in to the variations in elastic properties Buehler Ltd., Lake Bluff, IL.). Between the polishing and measurements,
and porosity in the proximal femur with respect to age. Furthermore, an ultrasound cleaner and a soft paint brush were used to remove debris
the relationship between tissue elastic properties and porosity has and diamond suspension from the specimen.
not been examined using bones harvested from the most severe frac-
ture site [27]. The present study aims to fill this gap in our knowledge. Scanning acoustic microscopy
We collected data on the regional distribution in the acoustic imped-
ance in tissue along the cross-section of the femoral neck and shaft A custom built quantitative scanning acoustic microscope was
and investigated the variation of tissue acoustic impedance values used [32]. Briefly, the SAM consisted of a three-axis high-precision
with age. Furthermore, microstructure and cortical porosity were an- scanning stage, a 200 MHz pulser–receiver (Panametrics 5900PR,
alyzed to determine possible variations with respect to aging and be- Panametrics-NDT, Waltham, MA), and a 400 MS/s 12-bit A/D-board
tween skeletal sites. This is essential information if one wishes to (CompuScope 12400, GaGe, Lockport, IL). All components were con-
estimate diagnostic determinants of bone fractures in elderly. trolled by custom-made software (SAM200Ex, Q-BAM, Halle, Germany).
Measurements were conducted in a temperature controlled tank
Materials and methods containing 25 °C distilled and degassed water using a 50 MHz spherical
focused transducer (V605/60°, −6 dB bandwidth 26–64 MHz, Valpey
Twenty-one human proximal femurs were obtained from male ca- Fisher, Hopkinton, USA) with a focal length of 5.2 mm and beam
davers (n = 21, aged 47.1 ± 17.8 years, range 17–82 years) at Kuopio diameter at focus of 23 μm [14,20,32].
University Hospital. Ethical approval for collection of samples was The incidence angle between the ultrasound pulse and the water-
granted by the National Authority for Medicolegal Affairs (permission bone interface was manually adjusted by using a custom made tilting
number: 5783/04/044/07). The cadavers were classified into three table. In order to ensure the perpendicular alignment and focusing of
groups, young (b 40 years, n = 8), middle aged (40–60 years, n = 9) the transducer on the bone surface, a scout scan was conducted over
and old (>60 years, n = 4). The cadavers had no pre-existing conditions the bone area before the actual measurements. After adjusting the
that might have affected bone metabolism (Table 1). time of flight between the bone surface and the transducer to be
After soft tissue removal, the proximal femurs were stored in a constant within the region of interest (ROI), optimal perpendicularity
freezer (−20 °C) until extracting the test samples. It has been reported between the incident ultrasound pulse and the water-bone interface
 M.K.H. Malo et al. / Bone 53 (2013) 451–458 453

Fig. 1. Cross-sections extracted from the femoral neck and shaft are indicated with boxes delineated with a dashed line in the X-ray image (a). The figures (b) and (c) illustrate
acoustic impedance maps obtained with SAM (50 MHz, pixel size 16 μm × 16 μm) for cross-sectional samples of the femoral neck and proximal shaft, respectively, of a 49 year-old
male cadaver. The magnified image of the shaft (d) shows that the pores can be distinguished from the mineralized tissue matrix. The interface between the cortical and trabecular
bone surface, identified by an experienced histologist, is indicated with a red line (b, c). The four regions of interests are indicated by solid yellow lines. The scale bars are specific for
the subimage in which they occur.

was reached. C-scan images were acquired along the x-y-plane using a
scan increment of 16 μm.
All signals were band-pass filtered using a zero-phase type II
Chebyshev filter with cutoff frequencies of 5 and 90 MHz. The ampli-
tude of the reflected signal was determined as the maximum of
the Hilbert-transformed (envelope) signal and the acoustic impedance
(Z) was calculated by using a calibration curve [19]. In order to create
the calibration curve, homogenous elastic materials with known acous-
tic impedances [Titanium, Suprasil® (Heraeus Quarzglas GmbH & Co.,
KG, Hanau, Germany), polymethylmethacrylate (PMMA), polystyrene
and polymethylpentene (TPX®)] were used as reference phantoms.
Calibration measurements for all reference phantoms were performed
before, in the middle of and after the measurement series. The titanium
and PMMA phantoms were measured daily to ensure the stability of the
system. The procedure of acoustic impedance measurement has been
extensively described elsewhere and is based on the estimation of
reflection amplitudes using the time of flight defocus correction
[14,19,22,32].
Determination of acoustic impedance, elastic coefficient and microstructure
In order to assess the spatial variations of elastic and microstructural
properties within each sample, four ROIs were defined for the neck and
shaft: superior, inferior, anterior, posterior and medial, lateral, anterior,
posterior, respectively (Fig. 1). Within each ROI, the endocortical
boundary between cortical and trabecular bone was determined by an
experienced histomorphometrist (HK). The cortical bone was demar-
cated by the following criteria. 1) Cortical bone includes Haversian
canals. 2) If the size of a pore is less than twice the average size of
pores at the region, then the pore is included in the cortex. 3) For the
larger pores, the endocortical boundary is set to split the pore. 4) If
the diameter of the pore is smaller than the distance from the pore to
454 M.K.H. Malo et al. / Bone 53 (2013) 451–458

Table 2 Results
Summary of the investigated parameters, their abbreviations and units.

Abbreviations Calculated parameter Unit Elastic coefficients

Ct.Th Cortical thickness
Ct.Po Cortical porosity
Po.D Pore density
Tr.c33 Trabecular elastic coefficient (median)
Ct.c33 Cortex elastic coefficient (median)
Ct.Wba Cortical area/whole bone area
the endosteal region, then the pore is included in the cortex. In most
cases, cortical bone could easily be distinguished. For evaluation of the
relations in elastic properties between i) tissue types and ii) neck and
shaft regions, the measured values from the four quadrants were
pooled.
All parameters (Table 2) were calculated using custom-made MATLAB
based software (MATLAB® v.7.10; The MathWorks, Inc., Waltham, MA,
USA). Cortical porosity (Ct.Po) was determined as the ratio between the
total area of pores and cortical bone area [20]. Pores were defined to be
voids within the cortex. Pore density (Po.D) was calculated as the mean
pore number per square millimeter (1/mm2). The cross-sectional acoustic
impedance maps were binarized using an adaptive threshold procedure.
These masks were done in order to segment the bone tissue from pores
filled with the embedding material. The binary masks were processed
using a circular erosion operator having a radius of two pixels (32 μm).
With this procedure boundary pixels (in which the measured impedance
was artificially reduced by the partial volume effect) were excluded. Sub-
sequently, the spatial variation of the elastic coefficient could be analyzed
accurately [14]. The areal portion of cortical tissue (Ct.Wba) was deter-
[mm]
[%] The elastic coefficient of the cortical tissue was significantly (F =
[1/mm2] 112.7, p b 0.01, two-way ANOVA) higher than that of trabecular
[GPa] bone. Furthermore, the elastic coefficient was significantly (F = 39.7,
[GPa]
p b 0.01, two-way ANOVA) higher in the femoral shaft than in the neck
[%]
(Fig. 2a). There was no synergistic interaction between tissue type
and anatomic location in the elastic coefficient.
The values of elastic coefficient varied significantly (p b 0.05,
ANOVA) within the cross-sections of the neck and shaft. In the femoral
neck, the values of elastic coefficient for cortex in the posterior and in-
ferior quadrants were significantly (p b 0.05, Tukey Kramer) higher
than those in the superior quadrant. In the femoral shaft, the values of
elastic coefficient for cortex in the posterior and medial quadrants
were significantly (p b 0.05, Tukey Kramer) lower than those measured
in the anterior quadrant (Fig. 2b). The values of elastic coefficients in
trabecular tissue were lowest in superior and anterior quadrants in
the neck and shaft, respectively (Table 3).
The elastic coefficients of cortical and trabecular bone were signifi-
cantly correlated in the femoral neck (R2 =0. 56, pb 0.01) (Fig. 3a) and
shaft (R2 =0.40, pb 0.01). A significant correlation (R2 =0.63, pb 0.01)
a
42
40
38
c33 [GPa]

36
mined from the ratio between the cortical tissue area and the whole
bone area (containing both the trabecular and cortical tissue). 34
The relation between the acoustic impedance Z and the elastic
coefficient c has been studied earlier [23]. The elastic coefficient c 32
represents the apparent stiffness value of each measurement point
30
and was determined as follows [15]:
28
1:923
c ¼ 0:608Z ; ð1Þ Trabecular Cortex Trabecular Cortex
Neck Shaft

Asymmetry of the histograms of the cortical and trabecular bone b

elastic coefficients was evaluated by determining their skewness.
42
Skewness was calculated using the third central moment of the histo-
gram, as described in previous studies [33]. The skewness parameter 40
provides important information since it quantitatively describes the var-
Cortex c33 [GPa]

38
iation in the elastic coefficient within a cross-section of bone.
36

Statistical analyses 34

Statistical analyses were conducted with custom-made MATLAB
software utilizing the Statistics Toolbox and with SPSS software version
19 (SPSS® Inc., Chicago, IL, USA). Differences between the tissue types
(trabecular and cortex), and skeletal sites were evaluated by one-way
analysis of variance (ANOVA) followed by post hoc multiple compari-
son Tukey–Kramer tests. The interactions of tissue type and skeletal
sites on the variation of the elastic coefficient were investigated with
two-way ANOVA. The significances of difference in cortical porosity
and elastic coefficient between the quadrants of the cross-sections of
neck and shaft were evaluated by using the one-way ANOVA with
post hoc Tukey Kramer multiple comparison tests. The association be-
tween the impedance values and porosity was assessed by linear re-
gression analysis. All statistical results were considered significant for
p values less than 0.05.
32
30
28
Anterior Posterior Superior Inferior Anterior Posterior Medial Lateral
Neck Shaft
Fig. 2. Elastic coefficients of cortical and trabecular bone showed significant (pb 0.05,
Tukey Kramer) variation between the skeletal locations (a). Elastic coefficient of cortex
varied significantly (pb 0.05, Tukey Kramer) between quadrants extracted from femoral
neck and shaft (b). Each group contains values from all cadavers (n=21). The bars on
top of the figure indicate significant (pb 0.05, Tukey Kramer) differences between the
groups. The central line in the boxes indicates the median, the edges are the 25th and
75th percentiles, the whiskers represent the extreme values, and the crosses denote
outliers.
 M.K.H. Malo et al. / Bone 53 (2013) 451–458 455

Table 3
Microstructural and elastic parameters (mean±standard error) for all evaluated regions
a
40
(n=21). R² = 0.56
p < 0.01
Ct.Th Ct.Po [%] Po.D Tr.c33 Ct.c33 n = 21
[mm] [1/mm2] [GPa] [GPa] 38

Neck c33, cortex [GPa]

Neck Anterior 1.6 ± 0.1 12.0 ± 0.9 11.2 ± 0.5 31.3 ± 0.4 34.7 ± 0.5
Posterior 1.3 ± 0.1 9.1 ± 0.8 10.4 ± 0.4 31.5 ± 0.3 35.8 ± 0.5
Superior 1.7 ± 0.3 13.8 ± 1.2 11.0 ± 0.5 30.6 ± 0.5 33.7 ± 0.6 36
Inferior 2.8 ± 0.2 7.0 ± 0.6 11.6 ± 0.6 31.8 ± 0.4 35.9 ± 0.5

Shaft Anterior 5.0 ± 0.2 8.7 ± 0.5 14.0 ± 0.5 32.0 ± 0.3 38.6 ± 0.4 34
Posterior 4.6 ± 0.3 13.5 ± 1.2 13.0 ± 0.5 34.2 ± 0.4 36.3 ± 0.4
Medial 5.6 ± 0.2 8.5 ± 0.6 15.2 ± 0.6 33.6 ± 0.5 37.8 ± 0.5
Lateral 6.4 ± 0.2 9.8 ± 0.7 14.4 ± 0.5 34.7 ± 0.3 37.0 ± 0.3 32
Ct.Th = cortical thickness, Ct.Po = cortical porosity, Po.D = cortical pore density, c33 =
elastic coefficient, Tr = trabecular and Ct = cortical.
28 29 30 31 32 33 34

was found between the elastic coefficients of cortical and trabecular Neck c33, trabecular [GPa]
bone in the femoral neck and shaft (Fig. 3b).
The areal portion of cortex (Ct.Wba) was correlated significantly
b
(R 2 = 0.55, p b 0.01) with the values of elastic coefficient in cortex at 42 R² = 0.63
the posterior quadrant of the femoral shaft (Fig. 3c). p < 0.01
40 n = 42
Porosity

Shaft c33 [GPa]

38
The cortical porosity varied significantly (p b 0.05, ANOVA) within
the cross-sections of the neck and shaft (Fig. 4). The cortical porosity 36
in the superior quadrant of femoral neck was significantly (p b 0.05,
Tukey Kramer) higher than that detected in the posterior and inferior 34
quadrants. Moreover, cortical porosity in the anterior quadrant was
significantly (p b 0.05, Tukey Kramer) higher than the porosity in the 32
Trabecular
inferior quadrant (Fig. 4). In the femoral neck, cortical porosity and
Cortex
pore density were lowest and highest, respectively, in the inferior 30
quartile (Table 3). In the femoral shaft, cortical porosity in the poste- 28 30 32 34 36 38 40
rior quadrant was significantly (p b 0.05, Tukey Kramer) higher than Neck c33 [GPa]
the values in the anterior, medial and lateral quadrants (Fig. 4).
Cortical porosity correlated negatively and positively with the
c
100
elastic coefficient in the femoral neck (R 2 = 0.27, p b 0.05) and shaft R² = 0.55
Areal portion of cortex in shaft [%]

(R 2 = 0.2, p b 0.05), respectively. Furthermore, cortical porosity corre- p < 0.01

95 n = 21
lated positively and negatively (R 2 = 0.20 and R 2 = 0.28, p b 0.05) with
the thickness of the cortex in the femoral neck and shaft, respectively
90
(Table 4).

85
Age

Subject age was positively associated with the elastic coefficient of 80

trabecular and cortical bone in the femoral neck and shaft (R2 = 0.28 –
0.46, p b 0.01 – 0.05), and also with the cortical porosity in the shaft
(R2 = 0.43, p b 0.01) (Figs. 5a and b, Table 4). Moreover, a negative cor-
relation (R2 = 0.27, p b 0.05) was found between the age and cortical
thickness in the neck (Table 4).
A significant correlation (R2 = 0.22, p b 0.05) was found between the
c33 histogram skewness and age in shaft trabecular bone. No age depen-
dency was found in the shapes of the elastic coefficient histograms de-
termined for the other anatomical sites or tissue types.
Discussion
The increase in cortical porosity, the thinning of cortex and the re-
sorption of trabeculae cause a reduction in bone mass, which increases
the fracture risk as an individual ages. In addition, an age related in-
crease in the bone tissue elastic coefficient may be linked to increased
tissue brittleness, induced by either an increased tissue mineralization,
changes in the collagen-cross linking composition, or a combination
thereof [23]. Indeed, it has been shown that the mesoscale cortical
bone strength (i.e. a few millimeters in size) in the longitudinal direc-
tion is correlated with the longitudinal Young's modulus (R ~ 0.5) [34].
75
70
32 34 36 38 40
Posterior shaft c33, cortex [GPa]
Fig. 3. (a) Elastic coefficients (c33) of cortical and trabecular bone were significantly
correlated in femoral neck. (b) Furthermore, elastic coefficients of neck and shaft
were significantly correlated. (c) The increase in the portion of cortical bone in the
shaft was related to the increase in the values of the cortical bone elastic coefficient.
The dots and error bars in (a) and (b) represent the weighted means and standard
errors, respectively, of the mean values measured within the four quadrants of each
cross-section.
The mesoscale elastic properties are determined by the microstructure
(i.e. the porous network) and the elastic properties of the tissue [25].
Therefore, variations of these microscale properties may contribute to
the macroscopic mechanical properties and risk of fracture. We hypoth-
esized that two parameters, i.e., elastic coefficient and cortical porosity
would display significant variation with aging and between skeletal
456 M.K.H. Malo et al. / Bone 53 (2013) 451–458

35 a
42
30 R² = 0.42
40 p < 0.01
Cortex porosity [%]

25
38
R² = 0.38
p < 0.01
20 R² = 0.28

c33 [GPa]
36
p < 0.05
15 34
R² = 0.46
p < 0.01
10 32
Shaft cortex
30 Neck cortex
5
Shaft trabecular
28
Anterior Posterior Superior Inferior Anterior Posterior Medial Lateral
Neck trabecular
Neck Shaft 20 40 60 80 100
Age [years]
Fig. 4. Cortex porosity varied significantly (p b 0.05, Tukey Kramer) between quadrants
extracted from femoral neck and shaft. Each group contains values from all cadavers
(n = 21). The bars on top of the figure indicate significant (pb 0.05, Tukey Kramer) dif-
b
22
ferences between the groups. The central line in the boxes indicates the median, the
R² = 0.43
edges are the 25th and 75th percentiles, the whiskers represent the extreme values,
20 p < 0.01
and the crosses denote outliers.
n = 21

Shaft cortex porosity [%]

18
sites (neck vs. shaft) and tissue types (cortical vs. trabecular). In order to
16
evaluate these hypotheses, scanning acoustic microscopy (50 MHz)
measurements were made on cross-sectional samples from the femoral 14
neck and shaft of male cadavers of different ages without pre-existing
conditions that might have affected bone metabolism. Femoral neck 12
and proximal femoral shaft samples were obtained from typical
transcervical neck and subtrochanteric fracture sites, respectively. 10
It was found that the trabecular bone stiffness at the human femoral
8
neck and shaft was lower than the stiffness of the cortex (Fig. 2a). This
contradicts the findings of Turner et al. [28] who claimed that the elas- 20 30 40 50 60 70 80
ticity of trabecular tissue was similar to that of cortical tissue. In their Age [years]
study, the value of Young's modulus (E) for trabecular bone tissue was
intermediate between the values of the transverse Young's moduli Fig. 5. (a) Age accounted for a significant part of the variation in the values of elastic
(ET) and average E composed of ET and the longitudinal Young's moduli coefficient at different anatomical sites (n = 21 for each group). Moreover, cortical po-
(EL) measured from cortical bone tissue. However, if one compares the rosity in the shaft (b) increased with age.

EL and E values for the trabecular tissue, their results become consistent
with the present findings. Moreover, their study was limited by a differences in loading conditions, the values of elastic coefficient vary be-
smaller sample number (cross-sections from one human donor) and tween the femoral neck and shaft (Figs. 3a and b).
the low spatial resolution of the applied ultrasound transducer (60 μm). In the assessment of the spatial variations of elastic coefficient and
Despite the significant difference identified in the elastic coefficients be- microstructural properties, the samples were divided into four quad-
tween the tissue types, the elastic coefficients of trabecular and cortical rants in which the cortical and trabecular bone were individually ana-
bone tissue were positively correlated. This suggests that the general lyzed. The posterior and anterior quadrants of the femoral shaft
loading of the femur can have a similar effect on the stiffness of both cor- exhibited the lowest and the highest values of the elastic coefficient, re-
tical and trabecular bone matrix at the neck and shaft sites, via changes in spectively (Fig. 2b). This is in line with previous studies [21,35]. In the
the bone turnover and mineralization. However, due to the site specific femoral neck, the elastic coefficient was highest in the posterior and in-
ferior quadrants which are the areas bearing most of the loading during
standing and walking (Fig. 2b) [36]. Furthermore, we found that when
the areal portion of the cortex in the whole cross-section of the femoral
Table 4 shaft was high, the elastic coefficient in the cortex in the posterior quad-
Pearson correlation coefficients between cadaver characteristics, microstructure, and
rant was also high (Fig. 3c.). This indicates that the elastic coefficient of
elastic coefficients of human femoral neck and shaft (n = 42).
the bone tissue adapts to the existing loading conditions.
Femoral neck — Femoral shaft — Femoral shaft — Cortical and trabecular bone elastic coefficients were found to in-
cortical bone cortical bone trabecular bone crease with age (Fig. 5a.). This may be related to the age dependent
Ct.Th c33 Ct.Th c33 c33 changes in bone tissue mineral (i.e. crystallinity and mineral content)
Ct.Po 0.20⁎ 0.27⁎ 0.28⁎ 0.20⁎ 0.30⁎⁎ and organic phases (i.e. collagen cross-links and re-orientation)
Po.D 0.20⁎ 0.41⁎⁎ 0.23⁎ ns. ns. [37,38]. The importance of the organic phase is supported by the results
c33 0.26⁎ 1 ns. 1 1 of Wall et al., who reported that bone tensile strength decreased by 27%
Age 0.27⁎ 0.38⁎⁎ ns. 0.42⁎⁎ 0.28⁎
accompanied by a simultaneous decrease in density by only 3%, with
BMI ns. ns. 0.29⁎ ns. ns.
increasing age [39]. The increase in porosity (Fig. 5b) and the greater
ns.: not significant. thinning of cortex with age were observed in the present study. Thus,
Ct.Th = cortical thickness, c33 = elastic coefficient, Ct.Po = cortical porosity, Po.D =
cortical pore density, and BMI = body mass index.
the combination of decreased bone mass and more brittle bone tissue
⁎ p b 0.05. may be the key factors contributing to the increased susceptibility for
⁎⁎ p b 0.01. fractures during aging.
 M.K.H. Malo et al. / Bone 53 (2013) 451–458
The relation between cortical porosity and age has been studied at
the human femoral mid shaft and neck. These studies have consistently
reported an increase with age [40–46]. The highest porosity and the
strongest dependence on age were found in the posterior shaft of the
femur [18,45]. The results of the present study are in line with the pre-
vious publications [9,46–49]. Although cortical porosity is an important
determinant of cortical bone strength and toughness [9], its relation
with matrix elastic stiffness, as a function of age, has not been addressed
so far. In the femoral neck, porosity was positively correlated with cor-
tical thickness and negatively correlated with elastic coefficient. This is
evidence that the thick cortex of the femoral neck is porous and
trabecularizes with aging, contributing to decreased cortical thickness
while the dense core in the periosteum remains structurally unchanged
with its high elastic coefficient. In the femoral shaft, porosity was posi-
tively and negatively correlated with the values of elastic coefficient and
cortical width, respectively. This would result in a thin and porous cor-
tex having a high elastic coefficient, reflecting bone tissue adaptation to
the existing loading conditions.
As there has been increasing interest in conducting diagnostic ultra-
sound measurements directly in the most severe fracture sites, e.g., the
femoral neck, it is essential to understand the natural variation of tissue
elastic properties and porosity with aging and between different skeletal
sites [50–53]. The elastic properties and porosity affect the speed of
sound in bone; this is vital information when evaluating the cortical
thickness with ultrasound. Thus, if one neglected these factors and
used a fixed speed of sound for bone, this would lead to inaccuracies in
the measurement of cortical thickness [54–56]. The data obtained in
the present study could be transferred into computational models, e.g.,
for evaluating the acoustic wave propagation through the femoral neck
and shaft with realistic geometry, elastic coefficient, and porosity. More-
over, the present data may be supplemented with micro-computed to-
mography (μCT) data e.g. it can be used as an input for numerical
simulations of mechanical properties of proximal femur [25,57–59].
In the present study, the beam width was 23 μm in the focal plane.
This limits the resolution of the SAM images and therefore pores with
diameters less than 23 μm i.e. canaliculi, Volkmann canals and osteo-
cyte lacunae could not be resolved. However, they make only a minor
contribution to the total porosity of the cortical bone [27].
In the present study, only one elastic coefficient (c33) could be mea-
sured to describe elastic behavior of the bone tissue. In order to obtain
other elastic coefficients, samples should be cut into pieces and also
measured along transverse directions. However, this was not possible
during this study. The present study includes only bone samples from
men, since we did not have access to sufficient bone material from fe-
male cadavers. In the present study, cortical and trabecular tissue
areas were identified using the SAM impedance maps, instead of the
gold standard method, i.e., histomorphometry. Thus, the present meth-
od used for identification of tissue type may include some uncertainty.
Finally, we used PMMA embedded samples that are known to exhibit
a consistent increase of approximately 4% in the acoustic impedance
values, corresponding to about an 8% increase in the apparent stiffness
values in mineralized bone [60]. The PMMA within the Haversian chan-
nels have been excluded from the estimation of the elastic coefficient by
the segmentation procedure. However, the PMMA also infiltrates the
extracellular tissue matrix, cell lacunae and the canalicular network.
These pores cannot be resolved with the used transducer. The replace-
ment of water by the embedding material in the intra- and interfibrillar
spaces and in the micro-porous network (osteocyte lacunae and cana-
liculi) alters the elastic properties of the tissue. For example, embedded
bone tissue with a high degree of mineralization contains less PMMA
than tissue with a low degree of mineralization. The coefficient of vari-
ation of bone mineral density distribution (BMDD) in healthy adult sub-
jects has been reported to be in the order of 2.1% [61]. Wang and Ni
investigated the variability of osteocyte lacunar porosity in human cor-
tical tissue [62]. They observed a slight age-dependent decrease in oste-
ocyte lacunar porosity (mean and standard deviation: 1.85%± 0.19%,
457
0.0074% decrease per year). Based on the rule of mixtures, the variation
of matrix stiffness induced by the embedding can be estimated. In addi-
tion to a constant increase of approximately 2.8% for a normal volume
fraction of water (~27%), variations of tissue mineralization (~2%) and
osteocyte lacunar porosity within the investigated age range (~0.5%)
on the effective stiffness can be considered to have embedding related
effects in the order of ~0.3% and ~0.07%, respectively.
In conclusion, there was age-dependent variation in both the elastic
coefficient and the porosity of calcified tissue in human femoral neck
and shaft. Moreover, values of elastic coefficient and porosity were
found to vary within cross-sections of femoral neck and shaft. Further-
more, the properties of trabecular and cortical bone were interrelated in
the femoral neck and shaft. The age related increase in elastic coefficient
of bone tissue may contribute to increased fragility of elderly bone and
this should be taken in to account when using ultrasound in the diag-
nostics of fracture susceptibility.
Acknowledgments
Academy of Finland (128863), Finnish Funding Agency for Technolo-
gy and Innovation (TEKES 70015/05), Sigrid Juselius foundation, Kuopio
University Hospital (EVO project 5031342, PY120 musculoskeletal disor-
ders), Finnish Foundation for Technology Promotion, the National Grad-
uate School of Musculoskeletal Disorders and Biomaterials (TBDP), the
Strategic Funding of University of Eastern Finland, Emil Aaltonen Foun-
dation and Deutsche Forschungsgemeinschaft (grant Ra1380/7) and
BMBF (grant 01DS12027A) are acknowledged for financial support. Re-
search was conducted in the frameworks of the European Associated
Laboratory “Ultrasound based assessment of bone quality” (LEA-ULAB)
and the Baltic Sea Network “Quantitative Imaging of Functional Compe-
tence of the Musculoskeletal System” (QUIMUS).
Special laboratory technician Ritva Sormunen, SIB labs, University
of Eastern Finland is acknowledged for helping with sample prepara-
tion for acoustic microscopy.
References
[1] Kanis JA, Johansson H, Oden A, McCloskey EV. Assessment of fracture risk. Eur J
Radiol 2009;71:392–7.
[2] Parfitt AM. Age-related structural changes in trabecular and cortical bone: cellular
mechanisms and biomechanical consequences. Calcif Tissue Int 1984;36(Suppl.
1):S123–8.
[3] Melton III LJ. Hip fractures: a worldwide problem today and tomorrow. Bone
1993;14(Suppl. 1):S1–8.
[4] Kanis JA. Assessment of fracture risk and its application to screening for postmen-
opausal osteoporosis: synopsis of a WHO report. WHO Study Group. Osteoporos
Int 1994;4:368–81.
[5] Rikkonen T, Salovaara K, Sirola J, Karkkainen M, Tuppurainen M, Jurvelin J, et al.
Physical activity slows femoral bone loss but promotes wrist fractures in post-
menopausal women: a 15-year follow-up of the OSTPRE study. J Bone Miner
Res 2010;25:2332–40.
[6] Karkkainen M, Rikkonen T, Kroger H, Sirola J, Tuppurainen M, Salovaara K, et al.
Physical tests for patient selection for bone mineral density measurements in
postmenopausal women. Bone 2009;44:660–5.
[7] Karkkainen M, Rikkonen T, Kroger H, Sirola J, Tuppurainen M, Salovaara K, et al.
Association between functional capacity tests and fractures: an eight-year pro-
spective population-based cohort study. Osteoporos Int 2008;19:1203–10.
[8] Milovanovic P, Djonic D, Marshall RP, Hahn M, Nikolic S, Zivkovic V, et al.
Micro-structural basis for particular vulnerability of the superolateral neck trabec-
ular bone in the postmenopausal women with hip fractures. Bone 2012;50:63–8.
[9] Yeni YN, Norman TL. Fracture toughness of human femoral neck: effect of micro-
structure, composition, and age. Bone 2000;26:499–504.
[10] Raum K. Book paragraph in “Bone Quantitative Ultrasound”. London, New York:
Springer Dordrecht Heidelberg; 2011.
[11] Gluer CC. A new quality of bone ultrasound research. IEEE Trans Ultrason
Ferroelectr Freq Control 2008;55:1524–8.
[12] Isaksson H, Nagao S, Malkiewicz M, Julkunen P, Nowak R, Jurvelin JS. Precision of
nanoindentation protocols for measurement of viscoelasticity in cortical and tra-
becular bone. J Biomech 2010;43:2410–7.
[13] Rho JY, Zioupos P, Currey JD, Pharr GM. Microstructural elasticity and regional
heterogeneity in human femoral bone of various ages examined by nano-
indentation. J Biomech 2002;35:189–98.
[14] Lakshmanan S, Bodi A, Raum K. Assessment of anisotropic tissue elasticity of
cortical bone from high-resolution, angular acoustic measurements. IEEE Trans
Ultrason Ferroelectr Freq Control 2007;54:1560–70.
458 M.K.H. Malo et al. / Bone 53 (2013) 451–458
[15] Preininger B, Checa S, Molnar FL, Fratzl P, Duda GN, Raum K. Spatial-temporal
mapping of bone structural and elastic properties in a sheep model following
osteotomy. Ultrasound Med Biol 2011;37:474–83.
[16] Katz JL, Meunier A. Scanning acoustic microscope studies of the elastic properties
of osteons and osteon lamellae. J Biomech Eng 1993;115:543–8.
[17] Hofmann T, Heyroth F, Meinhard H, Franzel W, Raum K. Assessment of composi-
tion and anisotropic elastic properties of secondary osteon lamellae. J Biomech
2006;39:2282–94.
[18] Rohrbach D, Lakshmanan S, Peyrin F, Langer M, Gerisch A, Grimal Q, et al. Spatial
distribution of tissue level properties in a human femoral cortical bone. J Biomech
2012;45:2264–70.
[19] Raum K. Microelastic imaging of bone. IEEE Trans Ultrason Ferroelectr Freq Control
2008;55:1417–31.
[20] Raum K, Reisshauer J, Brandt J. Frequency and resolution dependence of the an-
isotropic impedance estimation in cortical bone using time-resolved scanning
acoustic microscopy. J Biomed Mater Res A 2004;71:430–8.
[21] Weiss S, Zimmerman MC, Harten RD, Alberta FG, Meunier A. The acoustic and
structural properties of the human femur. J Biomech Eng 1998;120:71–6.
[22] Raum K, Leguerney I, Chandelier F, Talmant M, Saied A, Peyrin F, et al.
Site-matched assessment of structural and tissue properties of cortical bone
using scanning acoustic microscopy and synchrotron radiation muCT. Phys Med
Biol 2006;51:733–46.
[23] Raum K, Cleveland RO, Peyrin F, Laugier P. Derivation of elastic stiffness from
site-matched mineral density and acoustic impedance maps. Phys Med Biol 2006;51:
747–58.
[24] Granke M, Grimal Q, Saied A, Nauleau P, Peyrin F, Laugier P. Change in porosity is
the major determinant of the variation of cortical bone elasticity at the millimeter
scale in aged women. Bone 2011;49:1020–6.
[25] Grimal Q, Raum K, Gerisch A, Laugier P. Derivation of the mesoscopic elasticity
tensor of cortical bone from quantitative impedance images at the micron scale.
Comput Methods Biomech Biomed Engin 2008;11:147–57.
[26] Bumrerraj S, Katz JL. Scanning acoustic microscopy study of human cortical and
trabecular bone. Ann Biomed Eng 2001;29:1034–42.
[27] Saied A, Raum K, Leguerney I, Laugier P. Spatial distribution of anisotropic acoustic
impedance assessed by time-resolved 50-MHz scanning acoustic microscopy and
its relation to porosity in human cortical bone. Bone 2008;43:187–94.
[28] Turner CH, Rho J, Takano Y, Tsui TY, Pharr GM. The elastic properties of trabecular
and cortical bone tissues are similar: results from two microscopic measurement
techniques. J Biomech 1999;32:437–41.
[29] Linde F, Sorensen HC. The effect of different storage methods on the mechanical
properties of trabecular bone. J Biomech 1993;26:1249–52.
[30] Sedlin ED, Hirsch C. Factors affecting the determination of the physical properties
of femoral cortical bone. Acta Orthop Scand 1966;37:29–48.
[31] van Haaren EH, van der Zwaard BC, van der Veen AJ, Heyligers IC, Wuisman PI,
Smit TH. Effect of long-term preservation on the mechanical properties of cortical
bone in goats. Acta Orthop 2008;79:708–16.
[32] Hube R, Mayr H, Hein W, Raum K. Prediction of biomechanical stability after callus
distraction by high resolution scanning acoustic microscopy. Ultrasound Med Biol
2006;32:1913–21.
[33] Preininger B, Hesse B, Rohrbach D, Varga P, Gerigk H, Langer M, et al. Histogram
feature-based classification improves differentiability of early bone healing stages
from micro-computed tomographic data. J Comput Assist Tomogr 2012;36:469–76.
[34] Grimal Q, Haupert S, Mitton D, Vastel L, Laugier P. Assessment of cortical bone
elasticity and strength: mechanical testing and ultrasound provide complementa-
ry data. Med Eng Phys 2009;31:1140–7.
[35] Espinoza Orias AA, Deuerling JM, Landrigan MD, Renaud JE, Roeder RK. Anatomic
variation in the elastic anisotropy of cortical bone tissue in the human femur.
J Mech Behav Biomed Mater 2009;2:255–63.
[36] Bergmann G, Deuretzbacher G, Heller M, Graichen F, Rohlmann A, Strauss J, et al. Hip
contact forces and gait patterns from routine activities. J Biomech 2001;34:859–71.
[37] Nyman JS, Reyes M, Wang X. Effect of ultrastructural changes on the toughness of
bone. Micron 2005;36:566–82.
[38] Zioupos P. Ageing human bone: factors affecting its biomechanical properties and
the role of collagen. J Biomater Appl 2001;15:187–229.
[39] Wall JC, Chatterji SK, Jeffery JW. Age-related changes in the density and tensile
strength of human femoral cortical bone. Calcif Tissue Int 1979;27:105–8.
[40] Bousson V, Bergot C, Meunier A, Barbot F, Parlier-Cuau C, Laval-Jeantet AM, et al.
CT of the middiaphyseal femur: cortical bone mineral density and relation to po-
rosity. Radiology 2000;217:179–87.
[41] Bousson V, Meunier A, Bergot C, Vicaut E, Rocha MA, Morais MH, et al. Distribution of
intracortical porosity in human midfemoral cortex by age and gender. J Bone Miner
Res 2001;16:1308–17.
[42] Cooper DM, Thomas CD, Clement JG, Turinsky AL, Sensen CW, Hallgrimsson B.
Age-dependent change in the 3D structure of cortical porosity at the human fem-
oral midshaft. Bone 2007;40:957–65.
[43] Feik SA, Thomas CD, Clement JG. Age-related changes in cortical porosity of the
midshaft of the human femur. J Anat 1997;191(Pt 3):407–16.
[44] Stein MS, Feik SA, Thomas CD, Clement JG, Wark JD. An automated analysis of
intracortical porosity in human femoral bone across age. J Bone Miner Res
1999;14:624–32.
[45] Thomas CD, Feik SA, Clement JG. Regional variation of intracortical porosity in the
midshaft of the human femur: age and sex differences. J Anat 2005;206:115–25.
[46] Chen H, Zhou X, Shoumura S, Emura S, Bunai Y. Age- and gender-dependent
changes in three-dimensional microstructure of cortical and trabecular bone at
the human femoral neck. Osteoporos Int 2010;21:627–36.
[47] Bell KL, Loveridge N, Power J, Garrahan N, Meggitt BF, Reeve J. Regional differ-
ences in cortical porosity in the fractured femoral neck. Bone 1999;24:57–64.
[48] Bell KL, Loveridge N, Reeve J, Thomas CD, Feik SA, Clement JG. Super-osteons
(remodeling clusters) in the cortex of the femoral shaft: influence of age and gender.
Anat Rec 2001;264:378–86.
[49] Chappard C, Bensalah S, Olivier C, Gouttenoire PJ, Marchadier A, Benhamou C, et al.
3Dcharacterization of pores in the cortical bone of human femur in the elderly at dif-
ferent locations as determined by synchrotron micro-computed tomography im-
ages. Osteoporos Int in press. http://dx.doi.org/10.1007/s00198-012-2044-4 [Epub
ahead of print].
[50] Barkmann R, Dencks S, Laugier P, Padilla F, Brixen K, Ryg J, et al. Femur ultrasound
(FemUS)—first clinical results on hip fracture discrimination and estimation of
femoral BMD. Osteoporos Int 2009;21:969–76.
[51] Barkmann R, Laugier P, Moser U, Dencks S, Klausner M, Padilla F, et al. A device for
in vivo measurements of quantitative ultrasound variables at the human proximal
femur. IEEE Trans Ultrason Ferroelectr Freq Control 2008;55:1197–204.
[52] Barkmann R, Laugier P, Moser U, Dencks S, Klausner M, Padilla F, et al. In vivo mea-
surements of ultrasound transmission through the human proximal femur. Ultra-
sound Med Biol 2008;34:1186–90.
[53] Karjalainen JP, Riekkinen O, Toyras J, Hakulinen M, Kroger H, Rikkonen T, et al.
Multi-site bone ultrasound measurements in elderly women with and without
previous hip fractures. Osteoporos Int 2012;23:1287–95. http://dx.doi.org/10.
1007/s00198-011-1682-2.
[54] Karjalainen J, Riekkinen O, Toyras J, Kroger H, Jurvelin J. Ultrasonic assessment of
cortical bone thickness in vitro and in vivo. IEEE Trans Ultrason Ferroelectr Freq
Control 2008;55:2191–7.
[55] Cowin SC, Cardoso L. Fabric dependence of wave propagation in anisotropic po-
rous media. Biomech Model Mechanobiol 2011;10:39–65.
[56] Raum K, Leguerney I, Chandelier F, Bossy E, Talmant M, Saied A, et al. Bone micro-
structure and elastic tissue properties are reflected in QUS axial transmission
measurements. Ultrasound Med Biol 2005;31:1225–35.
[57] Aula AS, Toyras J, Hakulinen MA, Jurvelin JS. Effect of bone marrow on acoustic
properties of trabecular bone—3D finite difference modeling study. Ultrasound
Med Biol 2009;35:308–18.
[58] Hakulinen MA, Day JS, Toyras J, Weinans H, Jurvelin JS. Ultrasonic charac-
terization of human trabecular bone microstructure. Phys Med Biol 2006;51:
1633–48.
[59] Grassi L, Schileo E, Taddei F, Zani L, Juszczyk M, Cristofolini L, et al. Accuracy of fi-
nite element predictions in sideways load configurations for the proximal human
femur. J Biomech 2012;45:394–9.
[60] Zimmerman MC, Prabhakar A, Chokshi BV, Budhwani N, Berndt H. The acoustic
properties of normal and imbedded bovine bone as measured by acoustic micros-
copy. J Biomed Mater Res 1994;28:931–8.
[61] Roschger P, Paschalis EP, Fratzl P, Klaushofer K. Bone mineralization density
distribution in health and disease. Bone 2008;42:456–66.
[62] Wang X, Ni Q. Determination of cortical bone porosity and pore size distribution
using a low field pulsed NMR approach. J Orthop Res 2003;21:312–9.