Vertebroplasty Abstact
Vertebroplasty Abstact
Vertebroplasty Abstact
Youssef El-Massoudi,
Peter Patka
BMC Musculoskeletal Disorders201112:34
DOI: 10.1186/1471-2474-12-34
Van Lieshout et al; licensee BioMed Central Ltd. 2011
Received: 11 October 2010
Accepted: 2 February 2011
Published: 2 February 2011
Abstract
Background
Many (artificial) bone substitute materials are currently available for use in
orthopaedic trauma surgery. Objective data on their biological and
biomechanical characteristics, which determine their clinical application, is
mostly lacking. The aim of this study was to investigate structural and in
vitro mechanical properties of nine bone substitute cements registered for use
in orthopaedic trauma surgery in the Netherlands.
Methods
Seven calcium phosphate cements (BoneSource, Calcibon, ChronOS,
Eurobone, HydroSet, Norian SRS, and Ostim), one calcium sulphate
cement (MIIG X3), and one bioactive glass cement (Cortoss) were tested.
Structural characteristics were measured by micro-CT scanning. Compression
strength and stiffness were determined following unconfined compression
tests.
Results
Each bone substitute had unique characteristics. Mean total porosity ranged
from 53% (Ostim) to 0.5% (Norian SRS). Mean pore size exceeded 100 m
only in Eurobone and Cortoss (162.2 107.1 m and 148.4 70.6 m,
respectively). However, 230 m pores were found in Calcibon, Norian SRS,
HydroSet, and MIIG X3. Connectivity density ranged from 27/cm3 for
HydroSet to 0.03/cm3 for Calcibon. The ultimate compression strength was
highest in Cortoss (47.32 MPa) and lowest in Ostim (0.24 MPa). Young's
Modulus was highest in Calcibon (790 MPa) and lowest in Ostim (6 MPa).
Conclusions
The bone substitutes tested display a wide range in structural properties and
compression strength, indicating that they will be suitable for different clinical
indications. The data outlined here will help surgeons to select the most
suitable products currently available for specific clinical indications.
Background
Treatment of bone defects is a continuous challenge in skeletal trauma and
orthopaedic trauma surgery. Bone graft represents the second most common
transplanted tissue, with blood being number one [1]. Worldwide, more than
2.2 million bone grafting procedures are performed annually for the repair of
bone defects in orthopaedic traumatology, neurosurgery, and dentistry [24].
Approximately 10% of all skeletal reconstructive surgical interventions require
bone grafting [4]. Large defects resulting from, among others, trauma,
infection, or tumor resection often do not heal spontaneously, and require
Methods
Sample preparation
Nine bone substitutes that were available as (injectable) paste were selected
for biomechanical testing; seven calcium phosphate cements, one calcium
sulphate and one bioactive glass (Table 1). The products were stored at room
temperature until use. Ten to 12 cylindrical test samples were prepared per
product using a custom-made Teflon mould (Dept. Experimental Medical
Product name
Producer
Calcium phosphate
BoneSource
Stryker Nederla
Calcibon
Biomet Europe
ChronOS Inject
Synthes, Inc
Eurobone
Surgical concep
HydroSet
Stryker Nederla
Norian SRS
Synthes, Inc
Ostim
Hereaus
Calcium sulphate
MIIG X3
Wright Medica
Bioactive glass
Cortoss
Orthovita, Inc
Figure 1
Micro-CT scanning
Architecture was determined using a micro-CT (Skyscan 1076, Kontich,
Belgium). The micro-CT was tuned at 70 kV and 140 A, with a resolution of 9
m. This setup was verified by scanning a Vitoss test sample with a known
porosity between 88 and 92% [21], which was indeed within this range (data
not shown). CT shadow projection images were converted into a three
dimensional reconstruction of cross-sectional images in bitmap files using the
volumetric reconstruction software (Nrecon software, Skyscan, Belgium).
Total, closed and open porosity, connectivity density, structure model index
(SMI) and pore size were calculated from these 3D reconstruction using the
CTAn software (SkyScan, Kontich, Belgium). Total porosity was defined as the
volume of all open plus closed pores as a percent of the total Volume Of
Interest (VOI) volume. Closed porosity represents the volume of the closed
pores as a percent of the total of solid plus closed pore volume within the VOI.
Open porosity is defined as the volume of open pores as a percent of the total
VOI volume. Connectivity density is the number of redundant connections
between trabecular structures per unit volume. The SMI indicates the relative
prevalence of rods and plates in a 3D structure. Pore size was defined as the
average thickness of the pores, similar to the definition of trabecular spacing
and thickness [22].
Biomechanical testing
The compression strength was determined using unconfined compression
tests. Upon five consecutive non-destructive preconditioning cycles, samples
were compressed at a velocity of 0.5 mm/min to fracture using a standard
compression-testing device (Lloyd Instruments, Fareham, UK). The resulting
Results
Sample characteristics
The average length and diameter were measured in order to check whether
the test samples size was as intended. Results are shown in Table 2. The
length ranged from 7.694 0.104 mm (mean SD) for Ostim to 8.365
0.085 mm for Eurobone. The diameter ranged from 3.650 0.103 mm for
Ostim to 3.992 0.047 for Calcibon. Both the length and diameter of
Ostim were statistically significantly less than the other products, implying
that the Ostim samples had slightly shrunken (p < 0.001, Mann-Whitney Utest).
Table 2
Average length, diameter and weight of the test samples
N
Length (mm)
Diameter (mm)
BoneSource
10
8.225 0.052
3.980 0.035
Calcibon
12
8.271 0.045
3.992 0.047
ChronOS
10
8.265 0.147
3.970 0.059
Eurobone
10
8.365 0.085
3.985 0.053
HydroSet
10
8.325 0.079
3.970 0.042
Norian SRS
10
8.180 0.079
3.915 0.034
Ostim
9*
7.694 0.104
3.650 0.103
MIIG X3
10
8.345 0.064
3.985 0.053
Cortoss
10
7.979 0.103
3.854 0.062
Figure 2
Densities of bone substitutes Densities of individual test samples were
calculated from their length, diameter and weight. Each dot represents an
individual test sample, and lines indicate the average value. The table below
the figure provides an overview of the statistical analysis of pairwise
comparisons (Student's T-test with Bonferroni correction). *, p < 0.05; **, p <
0.01; ***, p < 0.005; ns, not statistically significantly different. Grey boxes
represent the self-self combinations, which could not be tested.
Porosity and pore size
In order to gain insight into the porous structure of the bone substitute
materials, the porosity and pore sizes were calculated from micro-CT images.
Ostim was the only product that had a clear porous structure. The total
porosity (52.66 10.14%) was significantly higher than the porosity of all other
products (Figure 3A). The porosity of the other products diminished from 6.93
1.32% (ChronOS) to 0.48 0.15% for Norian SRS. As total porosity is
dictated by open as well as closed pores, the open porosity and closed
porosity were also determined. Open porosity was evident for Ostim (50.52
4.49%; Figure 3B), and diminished from 2.86 0.92% (ChronOS) to 0.22
0.75% for Calcibon. Closed porosity exceeded was highest for
ChronOS (3.59 0.41%) and HydroSet (2.66 0.49%), and lowest for
Ostim (0.43 0.32%), Norian SRS(0.33 0.13%), and MIIG X3 (0.29
0.07%; Figure 3C).
Figure 3
Porosity of bone substitutes Porosity of individual test samples was
determined upon Micro-CT-scanning as described in the Materials and
Methods. The total porosity (A), open porosity (B) and closed porosity (C)
were determined. Each dot represents an individual test sample, and lines
indicate the average value. The table below the figure shows the outcome of
the pairwise comparisons (Student's T-test with Bonferroni correction). *, p <
0.05; **, p < 0.01; ***, p < 0.005; ns, not statistically significantly different. Grey
boxes represent the self-self combinations, which could not be tested.
The porous structure of the bone substitute materials is determined by their
porosity and pore size. Only two products had a mean pore size that
exceeded 100 m, i.e., 162.2 107.1 m for Eurobone and 148.4 70.6 m
for Cortoss (Figure 4). Pore sizes of Norian SRS (47.2 21.9 m),
Calcibon (41.6 22.0 m) and BoneSource (33.4 6.2 m) were below 50
m.
Figure 4
Average pore sizes of bone substitutes Average pore sizes of individual
test samples were determined upon Micro-CT-scanning as described in the
Materials and Methods. Each dot represents an individual test sample, and
lines indicate the average value. The table below the figure shows the
outcome of the pairwise comparisons (Student's T-test with Bonferroni
correction). *, p < 0.05; **, p < 0.01; ***, p < 0.005; ns, not statistically
significantly different. Grey boxes represent the self-self combinations, which
could not be tested.
For each product the range in pore sizes is shown in Figures 5. Of all
products, BoneSource had the smallest pores. Over 95% of pores were
smaller than 60 m, of which approximately half were < 26.7 m. No pores >
100 m were found. This was also seen in Ostim, of which 95% of pores
were smaller than 85 m. Calcibon, Norian SRS and HydroSet incidentally
showed pores up to 230 m, however 95% were smaller than 125 m. Of the
CaPO4 products, ChronOS and Eurobone were the only two that contained
pores up to 500 m, with 95% of pores being smaller than 250 m and 330
m, respectively. The distribution of pore sizes of the CaSO4 MIIG X3
appeared similar as that of Norian SRS and, to a lesser extent, Calcibon.
However, with a maximum pore size of 250 m and 90% of pores being < 190
m, pores of MIIGX3 were relatively larger. The pore size frequency of
Cortoss deviated from that of the other products tested, as a large range of
pore sizes (25 to 300 m) were approximately equally present. In this
bioactive glass 95% of pores had sizes up to 390 m, although pores of 500
m were also found. Combining the data of total porosity and average pore
size implied that bone substitute materials provide a wide range of products.
Some had a high porosity with small pores (e.g., Ostim), and at the other side
of the spectrum products had a low porosity with large pores (e.g.,
Eurobone).
Figure 5
Figure 6
Connectivity density of bone substitutes The connectivity density of
individual test samples was determined upon Micro-CT-scanning as described
in the Materials and Methods. Each dot represents an individual test sample,
and lines indicate the average value. The table below the figure shows the
outcome of the pairwise comparisons (Student's T-test with Bonferroni
correction). *, p < 0.05; **, p < 0.01; ***, p < 0.005; ns, not statistically
significantly different. Grey boxes represent the self-self combinations, which
could not be tested.
Ostim was the only product with a positive structure model index (SMI) (0.125
1.165; Figure 7). For the other products, the SMI declined from -37.715
7.280 for ChronOS and -67.752 8.913 for HydroSet to -123.717 38.232
for Cortoss.
Figure 7
Structure model index of bone substitutes The structure model index of
individual test samples was determined upon Micro-CT-scanning as described
in the Materials and Methods. Each dot represents an individual test sample,
and lines indicate the average value. The table below the figure shows the
outcome of the pairwise comparisons (Student's T-test with Bonferroni
correction). *, p < 0.05; **, p < 0.01; ***, p < 0.005; ns, not statistically
significantly different. Grey boxes represent the self-self combinations, which
could not be tested.
Compression strength and Young's modulus
The compression strength of all products was determined using unconfined
compression tests. Cortoss had the highest ultimate compression strength
(47.32 20.34 MPa; see Figure 8). This was statistically significantly higher
than the strength of all other products. Next in order of diminishing strength
were Calcibon and Norian SRS (33.95 6.75 and 25.64 7.37 MPa,
respectively), which was statistically significantly higher than most other
products. ChronOS and Ostim had poor compression strengths (0.81 0.32
and 0.24 0.05 MPa, respectively).
Figure 8
Compression strength of bone substitutes The compression strength was
determined using unconfined compression tests as described in the Materials
and Methods. Each dot represents an individual test sample, and lines
indicate the average value. The table below the figure shows the outcome of
the pairwise comparisons (Student's T-test with Bonferroni correction). *, p <
0.05; **, p < 0.01; ***, p < 0.005; ns, not statistically significantly different. Grey
boxes represent the self-self combinations, which could not be tested.
Calcibon had the highest Young's modulus (790 132 MPa; Figure 9),
followed by Norian SRS and MIIG X3 (674 146 MPa and 665 154 MPa,
respectively). The Young's modulus of these three products was statistically
significantly higher than that of the other products. ChronOS and Ostim had
a very low Young's modulus (54 20 MPa and 6 3 MPa, respectively), which
was statistically significantly lower than all other products.
Figure 9
Young's modulus of bone substitutes The Young's modulus was
determined using unconfined compression tests as described in the Materials
and Methods. Each dot represents an individual test sample, and lines
indicate the average value. The table below the figure shows the outcome of
the pairwise comparisons (Student's T-test with Bonferroni correction). *, p <
0.05; **, p < 0.01; ***, p < 0.005; ns, not statistically significantly different. Grey
boxes represent the self-self combinations, which could not be tested.
Discussion
Osteoconductive porous biomaterials provide a scaffold for the ingrowth of
bone. With respect to pore size, microporosity (i.e., pores with a size < 5 m)
is considered important for the bioresorbability of the material [23], whereas
macroporosity (i.e., pores > 100 m) plays an important role in the
osteoconductivity. A large macroporosity (i.e., 400-600 m) facilitates
infiltration by fibrovascular tissue and revascularization, thereby allowing for
bone reconstruction. Investigations of bone ingrowth into porous materials
with varying pore size have led to the consensus that the optimal pore radius
for bone ingrowth is >50 m and perhaps as large as 150 m [2427]. Of the
bone substitute materials tested, Eurobone, Cortoss, and ChronOS could
be considered as truly osteoconductive in terms of pore size, as they contain
a considerable number of pores with sizes of up to 500 m. For ChronOS the
presence of pore sizes between 100 and 400 m have been shown before
[28, 29]. Pore sizes between 100-250 m are only marginally present in
Calcibon, HydroSet, MIIG X3, and Norian SRS. BoneSource and
Ostim do not contain pores with a size of at least 100 m, so based upon
this in vitromeasurement, these might not be considered as highly
osteoconductive based upon pore size alone. This is in agreement with
literature data available for BoneSource (2-50 m) [30] and Calcibon (<1 m)
[31].
The structure model index (SMI) indicates the relative prevalence of rods and
plates in a 3D structure. SMI involves a measurement of surface convexity.
Concave surfaces of enclosed cavities represent negative convexity to the
SMI parameter. SMI values of ideal plates, cylinders and spheres are 0, 3,
and 4, respectively. With a mean SMI value <0.2 Ostim appears to be mainly
composed of plates. It is known that products with a total porosity below 50%
often have a negative SMI. In this study, that was the case for eight out of nine
products. Overall, SMI is positively correlated with total porosity, open
porosity, and closed porosity (rp = 0.672, 0.645 and 0.358, respectively; p <
0.001), and negatively with compression strength and Young's Modulus (rp =
-0.679 and -0.638, respectively, p < 0.001).
Although faster ingrowth is favoured by a more porous and interconnected
structure, denser ceramics have better mechanical integrity [20, 24, 27]. For
example, an increase of the total porous volume from 10% to 20% can result
in a four-fold decrease in mechanical strength [27, 43, 44]. Of the bone
substitute products tested, this phenomenon is most pronounced for Ostim.
Ostim has the highest total porosity (mean ~53%), but has poor compressive
strength (mean 0.24 MPa) and Young's modulus (6 MPa). Calcibon and
Norian SRS, on the other hand, have low porosity (0.93% and 0.48%,
respectively), but display a relatively high compressive strength (33.9 MPa
and 25.6 MPa, respectively). Our data are in line with previous
measurements, which revealed a compression strength of 6.3-34 MPa for
BoneSource [45, 46], 35-55 MPa for Calcibon [34, 47], 14-24 MPa for
HydroSet [48], and 23-55 MPa for Norian SRS [4951]. For MIIG X3, an in
vivocompression strength of 0.6 MPa has been shown at 13 weeks follow up
in a canine fracture model [52]. This is lower than the 21.82 21.93 MPa
found in the current in vitro study, and is most likely due to a high degree of
biodegradation and resorption of the MIIG X3 graft, as calcium sulphates are
generally resorbed within 8-10 weeks. The 91-179 MPa as published for
Cortoss [53] is higher than we found. This may be due to the larger size of the
test samples (i.e., 8 7.5 100 mm) in the study by Boyd et al. [53]. As size
and shape of the tested samples as well as the test setup itself may influence
the outcome of the compression test, our data may allow for a more objective
comparison of strengths between the products.
Overall, compression strength was negatively correlated with total porosity
(rp = -0.424, p < 0.001), open porosity (rp = -0.399, p < 0.001), closed porosity
(rp = -0.412, p < 0.001), and connectivity density (rp = -0.220, p = 0.034) (data
not shown). Likewise, Young's modulus was negatively correlated with total,
open and closed porosity (rp = -0.573, -0.539 and -0.491, respectively, p <
0.001). As opposed to porosity, pore size was unrelated to the compression
strength (rp = 0.113, p = 0.281) or the Young's modulus (rp = -0.204, p = 0.050;
data not shown).
The compounds tested in the current study represent the major classes of
artificial bone grafts, i.e., calcium phosphates, calcium sulphate, and bioactive
glass. Although selected based upon their availability in the Netherlands, their
wide availability makes the data presented in this study generally relevant to
most countries. Synthetic calcium phosphate cements can be moulded to
irregularly shaped defects, or even injected via syringe before they harden in
situ. The two main forms of calcium phosphates currently used are betatricalcium phosphate (-TCP) and hydroxyapatite (HA), which can be used
separately or combined in composite cements. Due to a general lack of
macroporosity calcium phosphate cement degrades layer by layer from the
outside to the inside. HA-cements tested include such as Ostim and
HydroSethave a limited resorption rate. They are characterized by a high
porosity, but a relatively low compressive strength. Clinical indications for
Ostim include fractures of the tibia plateau [54, 55], calcaneus [54], and distal
radius [54, 56, 57]. There are currently no publications on clinical use of
HydroSet.
Conclusions
The nine bone substitutes studied each have their individual characteristics,
and provide orthopaedic trauma surgeons with a choice of products that
varies largely in architecture and strength. Only for Eurobone and
Cortoss the pore sizes exceed the 100 m that is regarded necessary for
proper osteoconduction. Biological and biomechanical characteristics of bone
substitutes determine their applicability and success rate. Therefore, the in
vivo behavior of these compounds (e.g., resorption rate and quality in bone
ingrowth) should be taken into account as well. In general, bioactive glass will
not resorb, and HA cements will remain in place for years. On the other hand,
calcium sulphate cements may disappear before bone ingrowth has taken
place. Calcium phosphate cements are generally densely packed and,
consequently, provide more mechanical strength. The data outlined here will
assist surgeons in selecting the most suitable product for specific clinical
indications. Further studies on their in vivo behavior are needed for
developing clinical guidelines for use of alternative bone substitute materials
in orthopaedic trauma surgery.
Declarations
Acknowledgements