D. KYTÝØ et al.: DEFORMATION BEHAVIOUR OF A NATURAL-SHAPED BONE SCAFFOLD
301–305
DEFORMATION BEHAVIOUR OF A NATURAL-SHAPED BONE
SCAFFOLD
OBNA[ANJE NARAVNO OBLIKOVANEGA OGRODJA KOSTI PRI
DEFORMACIJI
Daniel Kytýø1,2, Tomá{ Doktor1,2, Ondøej Jirou{ek1, Tomá{ Fíla1,2,
Petr Koudelka1,2, Petr Zlámal2
1Czech Technical University in Prague, Faculty of Transportation Sciences, Department of Mechanics and Materials, Konviktská 20,
110 00 Prague 1, Czech Republic
2Institute of Theoretical and Applied Mechanics, v.v.i., Academy of Sciences of the Czech Republic, Prosecká 76, 190 00 Prague 9,
Czech Republic
kytyr@fd.cvut.cz
Prejem rokopisa – received: 2014-08-11; sprejem za objavo – accepted for publication: 2015-05-07
doi:10.17222/mit.2014.190
The study aims at mechanical testing of an artificial bone structure in the form of a scaffold for the application in the repairs of
trabecular bones after wounds or degenerative diseases. Such artificial construct has to conform to many requirements including
biocompatibility, permeability properties and bone-integration characteristics. Recently, self-degradable bone scaffolds suitable
for natural-bone-tissue ingrowth optimized with respect to mechanical properties and body-fluid flow have been considered as
an alternative to allografts and autografts. Here, an analysis of deformation behaviour of a scaffold with a morphology identical
to the natural bone is the first step in this task. In this work, the geometry and morphology of scaffold specimens produced with
direct 3D printing were based on a 3D model derived from the X-ray-computed micro-tomography measurement of a real
trabecular bone. The geometrical model was upscaled four times in order to achieve the optimum ratio between its resolution
and the resolution of the 3D printer. For its biocompatibility and self-degradability, polylactic acid was used as the printing
material. The mechanical characteristics were obtained from a series of uniaxial compression tests, with an optical evaluation of
the strain field on the surfaces of the specimens. The acquired stress-strain curves were compared with the characteristics of a
real trabecular bone obtained with time-lapse microtomography measurements, evaluated with the digital volumetric correlation
method. The results show good correspondence of the stiffness values for both the natural and artificial bone specimens.
Keywords: bone scaffold, polylactic acid, additive manufacturing, compression loading, microtomography
Namen {tudije je mehansko preizku{anje umetnega ogrodja kosti za obnovo trabekularnih kosti po po{kodbah ali degenerativnih
boleznih. Tako umetno ogrodje mora ustrezati mnogim zahtevam, kot so biokompatibilnost, prepustnost in mo`nost vra{~anja
kostnega tkiva. Samo razgradljivo ogrodje kosti, primerno za naravno vra{~anje kostnega tkiva, optimirano glede na mehanske
lastnosti in toka telesnih teko~in, je bilo prou~evano kot nadomestek za presajanje tujega ali lastnega tkiva. Prvi korak pri tej
nalogi je analiza obna{anja ogrodja, z morfologijo podobno naravni kosti. V tem delu je bila geometrija in morfologija
vzor~nega ogrodja izdelana z neposrednim tridimenzionalnim tiskanjem, na osnovi tridimenzionalnega modela, dobljenega z
meritvami s pomo~jo rentgenske tomografije realne trabekularne kosti. Geometrijski model je bil pove~an {tirikrat, da bi dobili
optimalno razmerje med njegovo resolucijo in resolucijo tridimenzionalnega tiskalnika. Za biokompatibilnost in sâmo
razgradljivost, je bila za tiskanje uporabljena polilakti~na kislina. Mehanske zna~ilnosti so bile dobljene z vrsto enoosnih tla~nih
preizkusov in z opti~no oceno napetostnega polja na povr{ini vzorcev. Pridobljene krivulje napetost-raztezek so bile primerjane
z zna~ilnostmi realne trabekularne kosti, ki so bile dobljene z zaporednimi mikrotomografskimi meritvami in ocenjene z digital-
no volumetri~no metodo korelacije. Rezultati ka`ejo dobro ujemanje togosti obeh vzorcev, tako naravne kot umetne kosti.
Klju~ne besede: ogrodje kosti, polilakti~na kislina, tridimenzionalnega tiskanje, tla~no obremenjevanje, mikrotomografija
1 INTRODUCTION
The change in the lifestyle during the past decades
(the so-called modern lifestyle characteristic caused by a
lack of sufficient physical activities coupled with an
improper type of nutrition) coupled with an increasing
life expectancy in many countries caused a significant
increase in health-care costs. Globally more than 30 % of
women and 20 % of men in elderly age suffer from bone
disorders. The increase is expected to double by 2020,1
partially also due to an increased occurrence of obesity.
Here, bone-tissue engineering and specifically designed
implants with functionally graded properties represent a
modern approach to the bone-repair process2 with the
goal to create an implant that is anatomically and
functionally compatible with the surrounding tissue.3 In
the recent orthopaedic practice, repairs of defective
bones have been most commonly carried out using
autografts and allografts. Although used for many years,
natural grafts possess several limitations and potentials
for complications due to various influences including the
donor-site morbidity, a loss of bone inductive factors,
resorption during healing, anatomical variations, etc.4 To
overcome these problems, implants in the form of an
artificial bone represent an attractive alternative with a
significant, though not yet fully exploited potential. In
order to engineer such a synthetic scaffold with optimum
characteristics, several factors and parameters have to be
taken into account.5 Among such factors, the type of the
cancellous bone, mechanical properties, the geometry6
Materiali in tehnologije / Materials and technology 50 (2016) 3, 301–305 301
UDK 620.173:620.11:67.017 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(3)301(2016)
and the expected transport phenomena7 are the most
important for the design of a synthetic porous bone.
Thus, with respect to the selected application (i.e., the
location within the skeletal microarchitecture)8 stiffness
and permeability properties have to be properly selected
to precisely match the properties of the natural can-
cellous bone. Identification of the trabecular microstruc-
ture and comparison of deformation behaviour of the
natural bone with the synthetic replication is the first
step in the artificial-bone-structure optimization. A com-
bination of radiological imaging methods and additive
manufacturing was used for preparing a natural-shaped
bone scaffold. A set of uniaxial compression tests were
performed to obtain the scaffold loading response repre-
sented by stress-strain diagrams. The strain calculation
was based on image-registration techniques, digital
image correlation (DIC) for optical data and digital
volume correlation (DVC) for tomographic data.
2 MATERIALS AND METHODS
2.1 Natural bone sample
Trabecular bone tissue is a biological material con-
sisting of basic structural elements called trabeculae.
Individual trabeculae are joined into a network with a
high porosity (50–90 %) naturally adapted to distribute
and dissipate the energy from external loads. For the
purpose of this study, original natural bone samples were
prepared by drilling them from the caput femoris of
42-year and 72-year-old male donors. The samples with
a diameter of 5 mm and a height of approximately 10
mm were put into an ultrasonic bath to remove the
remaining bone marrow. Formaldehyde solution was
used to avoid biological degradation and to ensure
sample preservation. The upper and bottom parts of the
samples were embedded in a 2-component epoxy resin
and after the hardening, two plan-parallel baseplates
were cut with a precise saw. This preparation procedure
minimises the risk of a rapid collapse of the bone
samples under loading due to a crack damage of the
microstructure.
2.2 Microtomography measurement
In order to acquire a geometrically precise model of
the internal trabecular structure, a custom-designed
microtomography device (Figure 1) was employed. The
irradiation of the sample was performed using a micro-
focus X-ray tube with a high-resolution transmission
target (XWT 160, X-Ray Worx, Germany). Radiograms
were acquired using a large (410 mm × 410 mm) X-ray
flat-panel scintillating detector (XRD 1622, PerkinElmer
Inc., USA) with an effective resolution of 2048 pixels ×
2048 pixels with a 200 μm pitch. A custom-designed
loading device composed of a stiff frame with a low
absorption of X-rays was designed specially for the
microtomography under loading. Force was gradually
applied on the specimens using the translation stage
(7T173-20, Standa Ltd., Lithuania) with a 2 μm tracking
accuracy and the applied force was measured using a
small-scale force transducer (U9b, HBM, Germany) with
a nominal capacity of 500 N. Before the loading proce-
dure, a sample was scanned in 720 projections (acquisi-
tion 2 × 0.5 s) with an angular step of 0.5° to obtain a
detailed geometrical model of the microstructure. Then a
time-lapse tomography (tomography under loading)
measurement was performed to obtain the information
about the spatial strain distribution during the loading.9
The specimen was incrementally loaded with 1 %
increment up to the total deformation of 6 %. After each
loading step, microtomography was performed to capture
the deformed microstructure. Acquisitions with 180 pro-
jections at 2° were used for this purpose to reduce the
computational costs of the reconstruction procedure. The
spatial strain distribution was assessed utilizing DVC,
which is an extension of the image-registration techni-
ques to three dimensions.10
Linearization of the attenuation range (a beam-
hardening correction) was applied to reduce the noise
and improve the contrast of individual projections.11 Due
to a high porosity of the samples and a small thickness of
the trabeculae (approximately 200 μm), the cone-beam
reconstruction algorithm12 was used to eliminate the
distortion of the reconstructed data caused by the
divergent nature of the X-ray beam.
2.3 Polylactic acid
Polylactic acid (C3H4O2)n (PLA) is a biodegradable
thermoplastic polyester derived from biomass and pro-
duced from the starch of various crop plants (e.g., corn,
cereals, potatoes, etc.). The material properties of PLA
are comparable with those of synthetic plastics, but they
are achieved at significantly lower energetic require-
ments.13 PLA is highly biocompatible14 and suitable for
the use in 3D printers making an attractive solution for
endoprosthesis.
D. KYTÝØ et al.: DEFORMATION BEHAVIOUR OF A NATURAL-SHAPED BONE SCAFFOLD
302 Materiali in tehnologije / Materials and technology 50 (2016) 3, 301–305
Figure 1: Microtomography set-up: 1) X-ray source, 2) detector, 3)
rotary stage, 4) loading device, 5) load-cell controller, 6) linear stage,
7) anti-vibration table and 8) cooling
Slika 1: Naprava za mikrotomografijo: 1) vir rentgenskih `arkov, 2)
detektor, 3) nosilec za rotiranje, 4) naprava za obremenjevanje, 5) nad-
zor pri obremenjevanju, 6) linearni nosilec, 7) antivibracijska mizica
in 8) hlajenje
2.4 PLA bulk testing
The production of the samples was carried out using
a Profi3Dmaker (Aroja, Czech Republic) 3D printing
device equipped with a 200 μm printing nozzle, operated
in the rapid additive-manufacturing mode with a slice
resolution of 250 μm. The material properties of the PLA
printer filament are not guaranteed by the producer and
the influence of the printing process on the material
characteristics is unknown; therefore, cylinder samples
with dimensions of 20 mm × 28 mm (diameter, height)
were tested in different printing modes to choose the
ideal printing set-up for the trabecular-bone replica
manufacturing. The samples were produced at 100 %, 70
% and 50 % filling levels. With all the filling levels, the
products consisted of a 500 μm shell and a core with a
defined filling amount. In the case of a reduced filling
amount, the core was created using a random-oriented
fibre meshwork with a defined overall porosity. To verify
the declared values of the filling level, the samples were
weighed with a laboratory scale.
2.5 Bone scaffold development
The reconstructed tomographic-image data were sub-
jected to the standard post-processing methods (includ-
ing thresholding and identification of the connected
components) using custom segmentation and modelling
software to obtain binary spatial-image data without
isolated fragments. Then the marching cube algorithm15
was used to extract the polygonal mesh of an isosurface
from the three-dimensional scalar-image data (as
depicted in Figure 2). Subsequently, this polygonal mesh
was carefully smoothed and decimated in an iterative
manner to obtain a surface suitable for 3D printing.
Because of the technical parameters of the available 3D
printer, the model was upscaled three times to ensure a
proper geometry of the printed replica.
2.6 Compression tests
The compression test of both bulk and porous speci-
mens was performed in a displacement-controlled
loading mode using a custom-designed uniaxial loading
device with a loading capacity of up to 2 kN. The
load-bearing frame was designed as an open cylinder
with expanded ends mounted on a stiff metallic plate
manufactured from polyamide-imide thermoplastic. The
displacement of the loading platens was controlled by
stepper motor SX17-1705 (Microcon, Czech Republic)
attached to a CPU 17A 100 harmonic drive (Harmonic
Drive, USA) with a transmission ratio of 250 : 1 leading
to an accuracy of displacement in the order of micro-
meters. High-accuracy load cell U9b (HBM, Germany)
was connected to an OM 502T (Orbit Merret, Czech
Republic) programmable indicator. The samples were
loaded up to a 20 % deformation at a constant loading
rate of 20 μm s–1. Strains were derived from optically
measured deformations, evaluated with the DIC algo-
rithm. For this purpose, images of the loaded specimens
were acquired with a high-resolution CCD camera
(Manta G-504B, AVT, Germany) attached to telecentric
zoom lens TCZR 072 (Opto Engineering, Italy). An
in-house software based on the GNU/Linux real-time
operation software and LinuxCNC open-source project
was used to control the experiments. For the DIC pro-
cedure, a custom Matlab tool16 based on the Lucas-Kana-
de algorithm17 was used.
3 RESULTS
The material properties of bulk PLA were measured
based on the compression tests of the PLA solid cylin-
drical samples at different levels of filling. Stress-strain
curves obtained from the force record and optically
measured strain are depicted in Figure 3. From the slope
of the linear parts of the stress-strain diagrams, Young’s
moduli presented in Table 1 were estimated.
D. KYTÝØ et al.: DEFORMATION BEHAVIOUR OF A NATURAL-SHAPED BONE SCAFFOLD
Materiali in tehnologije / Materials and technology 50 (2016) 3, 301–305 303
Figure 3: Stress-strain diagram of the PLA bulk material with
different levels of filling
Slika 3: Obremenitveni diagram PLA materiala z razli~nimi stopnjami
polnjenja
Figure 2: Model of trabecular-bone microstructure used for additive
manufacturing
Slika 2: Model mikrostrukture trabekularne kosti, uporabljen pri tri-
dimenzionalnem tiskanju
Table 1: Properties of PLA used for additive manufacturing
Tabela 1: Lastnosti PLA, ki se uporabljajo pri tridimenzionalnem
tiskanju
Sample Filling (%) Weight (g) E (GPa)
1 100 15.300 2.396
2 70 11.937 1.864
3 50 9.160 1.380
Because of a low PLA elastic modulus (compared to
the bone tissue) and a rod- or shell-like shape of the
replicated inner structure, the PLA scaffold was manu-
factured only in the full filling mode. Four identical PLA
specimens representing the natural-bone-shaped
upscaled scaffold were used for the compression test. For
the optical measurement of the deformation of the com-
plex sample surface, the data acquired from four diffe-
rent loading scenes (rotated by 90°) were evaluated. A
sample collapsed after the compression test is depicted in
Figure 4.
Deformation behaviour of the PLA models of the
trabecular structure in comparison with the stress-strain
curves of the natural bone structure is depicted in Figure
5. Good agreement between the elastic parts of both
materials was observed. The sample with a natural bone
volume fraction (bone volume/total volume = 25 %)
exhibits a stiffness similar to the real-bone structure.
Therefore, PLA is suitable for other bone-scaffold deve-
lopments based on regular cells.
4 CONCLUSION
Compression tests of the natural-shaped additive-
manufactured PLA samples of the trabecular-bone struc-
ture were performed to assess deformation behaviour of
the homogenous solid phase of the synthetic material.
The results will be used for designing an artificial bone
structure with effective mechanical properties, close to
the ones of a real, healthy bone. A low bone volume/total
volume ratio of the real-bone structure provides a suffi-
cient reserve for increasing the strength of the scaffold
by increasing its relative density. A higher cross-section
of the basic structure elements (which would still enable
the tissue ingrowth) may reduce the required resolution
of manufacturing devices. Additive manufacturing of
biodegradable complex structures is a promising way for
the bone-scaffold development. Based on these findings,
the artificial bone structure used for the replacements of
trabecular bones will be optimized with respect to
structural, mechanical and permeability properties.
Acknowledgements
The research was supported by the Grant Agency of
the Czech Technical University in Prague (grant No.
SGS15/225/OHK2/3T/16) and by institutional support
RVO: 68378297.
5 REFERENCES
1 U.S. Department of Health and Human Services, Bone Health and
Osteoporosis: A Report of the Surgeon General, Rockville 2004
2 H. L. M. Bao, E. Y. Teo, M. S. K. Chong, Y. Liu, M. Choolani, J. K.
Y. Chan, Advances in Bone Tissue Engineering, Regenerative
Medicine and Tissue Engineering, InTech, Rijeka 2013,
doi:10.5772/55916
3 H. Razi, S. Checa, K. Schaser, G. N. Duda, Shaping scaffold struc-
tures in rapid manufacturing implants: A modeling approach toward
mechano-biologically optimized configurations for large bone defect,
Journal of Biomedical Materials Research, Part B: Applied
Biomaterials, 100 (2012) 7, 1736–1745, doi:10.1002/jbm.b.32740
4 A. R. Vaccaro, K. Chiba, J. G. Heller, T. C. Patel, J. S. Thalgott, E.
Truumees, J. S. Fischgrund, M. R. Craig, S. C. Berta, J. C. Wang,
Bone grafting alternatives in spinal surgery, Spine Journal, 2 (2002)
3, 206–215, doi:10.1016/S1529-9430(02)00180-8
5 A. Gantar, L. P. Da Silva, J. M. Oliveira, A. P. Marques, V. M. Corre-
lo, S. Novak, R. L. Reis, Nanoparticulate bioactive-glass-reinforced
gellan-gum hydrogels for bone-tissue engineering, Materials Science
and Engineering C, 43 (2014), 27–36, doi:10.1016/j.msec.2014.
06.045
6 V. P. W. Shim, L. M. Yang, J. F. Liu, V. S. Lee, Characterisation of
the dynamic compressive mechanical properties of cancellous bone
from the human cervical spine, International Journal of Impact Engi-
neering, 32 (2005) 1–4, 525–540, doi:10.1016/j.ijimpeng.2005.
03.006
D. KYTÝØ et al.: DEFORMATION BEHAVIOUR OF A NATURAL-SHAPED BONE SCAFFOLD
304 Materiali in tehnologije / Materials and technology 50 (2016) 3, 301–305
Figure 5: Stress-strain diagrams for the PLA-additive-manufactured
and natural bone
Slika 5: Diagram napetost-raztezek za 3D tiskano in naravno kost
Figure 4: Example of a PLA scaffold collapsed after the compression
test
Slika 4: Primer poru{enega PLA kostnega ogrodja po tla~nem preiz-
kusu
7 S. S. Kohles, J. B. Roberts, M. L. Upton, C. G. Wilson, L. J.
Bonassar, A. L. Schlichting, Direct perfusion measurements of
cancellous bone anisotropic permeability, Journal of Biomechanics,
34 (2001) 9, 1197–1202, doi:10.1016/S0021-9290(01)00082-3
8 E. F. Morgan, H. H. Bayraktar, T. M. Keavenyemail, Trabecular bone
modulus-density relationships depend on anatomic site, Journal of
Biomechanics, 36 (2003) 7, 897–904, doi:10.1016/S0021-9290
(03)00071-X
9 O. Jirousek, P. Zlamal, D. Kytyr, M. Kroupa, Strain analysis of trabe-
cular bone using time-resolved X-ray microtomography, Nuclear
Instruments and Methods in Physics Research, Section A: Accele-
rators, Spectrometers, Detectors and Associated Equipment, 633
(2011) 1, 148–151, doi:10.1016/j.nima.2010.06.151
10 O. Jirousek, I. Jandejsek, D. Vavrik, Evaluation of strain field in mi-
crostructures using micro-CT and digital volume correlation, Journal
of Instrumentation, 6 (2011) 1, C01039, doi:10.1088/1748-0221/6/
01/C01039
11 D. Vavrik, J. Jakubek, Radiogram enhancement and linearization
using the beam hardening correction method, Nuclear Instruments
and Methods in Physics Research Section A: Accelerators,
Spectrometers, Detectors and Associated Equipment, 607 (2009) 1,
212–214, doi:10.1016/j.nima.2009.03.156
12 L. A. Feldkamp, L. C. Davis, J. W. Kress, Practical cone-beam
algorithm, Journal of the Optical Society of America A, 1 (1984) 6,
612–619
13 A. P. Mathew, K. Oksman, M. Sain, Mechanical properties of bio-
degradable composites from poly lactic acid (PLA) and micro-
crystalline cellulose (MCC), Journal of Applied Polymer Science, 97
(2005) 5, 2014–2025, doi:10.1002/app.21779
14 J. M. Anderson, M. S. Shive, Biodegradation and biocompatibility of
PLA and PLGA microspheres, Advanced Drug Delivery Reviews, 28
(1997) 1, 5–24, doi:10.1016/S0169-409X(97)00048-3
15 W. E. Lorensen, H. E. Cline, Marching cubes: A high resolution 3D
surface construction algorithm, ACM SIGGRAPH Computer
Graphics, 21 (1987) 1, 163–169
16 I. Jandejsek, J. Valach, D. Vavrik, Optimization and Calibration of
Digital Image Correlation method, Proc. of Experimental Stress
Analysis, Olomouc 2010, 121–126
17 B. D. Lucas, T. Kanade, An iterative image registration technique
with an application to stereo vision, Proc. of the 7th International
Joint Conference on Artificial Intelligence – Volume 2, San Fran-
cisco 1981, 674–679
D. KYTÝØ et al.: DEFORMATION BEHAVIOUR OF A NATURAL-SHAPED BONE SCAFFOLD
Materiali in tehnologije / Materials and technology 50 (2016) 3, 301–305 305