( PDF ) Rev Osteoporos Metab Miner. 2012; 4 (4): 127-32

Guede D1,2, Pereiro I3, Solla E3, Serra J3, López-Peña M4, Muñoz F4, González-Cantalapiedra A4, Caeiro JR2,5, González P3
1 Trabeculae – Empresa de Base Tecnológica, S.L. – Ourense
2 Red Temática de Investigación en Envejecimiento y Fragilidad (RETICEF) – Instituto de Salud Carlos III – Ministerio de Economía y Competitividad – Madrid
3 Grupo de Nuevos Materiales – Departamento de Física Aplicada – Universidad de Vigo – Pontevedra
4 Departamento de Ciencias Clínicas Veterinarias – Facultad de Veterinaria – Universidad de Santiago de Compostela – Lugo
5 Servicio de Cirugía Ortopédica y Traumatología – Complejo Hospitalario Universitario de Santiago de Compostela – A Coruña


Background: The new generation of materials for implants should imitate the hierarchical structures found in nature. Bio-inspired silicon carbide ceramic (bioSiC) is a ceramic produced from wood, which has a similar structure to bone, with a unique property of interconnected porosity, which allows the internal growth of tissue and favours angiogenesis.
Objectives: To evaluate the biocompatibility and osteointegration of bioSiC in femoral bone defects in an experimental model in rabbits.
Material and methods: 36 cylinders of bioSiC were obtained through pyrolysis of sapelli wood and infiltration with molten silicon of the resulting carbon preform. Eighteen cylinders were coated with Si-HA by pulsed laser deposition. The cylinders were implanted in femoral condyles of rabbits which were sacrificed at 1, 4 or 12 weeks. The samples were analysed histologically using an optical microscope and computerised microtomography to assess bone growth.
Results: The bioSiC implants showed good osteointegration, there being both outward growth (ongrowth) and inward growth (ingrowth). At 4 weeks from implantation the integration was almost complete, with no difference from that seen at at 12 weeks. The coating did not improve the value of any parameter with respect to the non-coated implants.
Conclusions: BioSiC ceramics produced from porous wood have good osteointegration and their interconnected porosity is colonised by bone tissue. In addition, they do not require the bioactivity of a coating to improve the apposition of neoformed bone. BioSiC stands as a material to be taken into account in biomedical applications.
Keywords: silicon carbide, bone substitutes, osteointegration, histological techniques, X-ray computerised microtomography.
Traditionally, bone defects have been treated by autologous (coming from the patient themselves), allogenic (coming from another individual of the same species) or xenogenic (with bone obtained from another species) tissue transplants, or even by implanting substitute materials. The autografts, although they have a shown a high percentage rate of success, are limited by the quantity of tissue which may be extracted and by the morbidity of the extraction site1. On their part, allografts and xenografts (which are much less often used) may cause immune reactions and transmit pathogens to the patient. With regard to synthetic bone substitutes, all of these enjoy a number of advantages over bone grafts, such as their unlimited availability or their ease of sterilisation. However, in spite if their great variety, the main problem with all of them is their limited ability to provide mechanical support in the affected zone2. Current demands on the engineering of tissues for the regeneration of bone tissue raises new challenges in material sciences and biomedical engineering. Arising from this is a trend towards technology based on biomimetics for the development of new biomaterials, which are inspired by models and biostructures which nature offers, and which have been perfected and optimised through the evolutionary process.
Bioinspired ceramic materials, based on natural precursors such as wood, plants or algae, bring together a number of the characteristics required of this new generation of biomaterials: good biomechanical properties, lightness, toughness, and with a hierarchical and interconnected porous structure which assimilates with bone tissue to encourage the growth of this tissue in its interior, as well as angiogenesis, which means to say the osteointegration of the material into the host tissue3. Bioinspired silicon carbide ceramics (BioSiC) bring together all the biomechanical requirements sufficient for their use as bone substitutes. These ceramics may be made from a variety natural structures4,5. The technological fabrication process is based on the ceramicisation of the natural precursors, obtaining as a final result a piece of silicon carbide with an interconnected porosity which maintains its hierarchical structure in terms of the size and distribution of the pores of the original material.
Mechanical trials indicate that the properties of these materials may be made to measure by the selection of the appropriate vegetal precursor from which they are made, depending on the characteristics of the type of bone which is being attempted to be repaired, and by modifying the experimental conditions of the fabrication process. Thus, it is notable that, as with bone, bioSiC has anisotropic mechanical properties, with values which compare favourably with those of bone. For example, the bioSiC obtained using sapelli wood (Entandrophragma cylindricum) has values of resistance to longitudinal compression (the growth axis of the tree) of 210±20 to 1.160±100 MPa, and a radial direction from 120±10 to 430±50 MPa as a function of the quantity of Si with which it has been infiltrated6,7. The resistance to compression of specimens of human cortical bone vary between 167 and 215 MPa8.
The biocompatibility of a material refers to its capacity to function as a substrate which will support appropriate cell activity, facilitating systems of cellular and mechanical signalling, with the aim of optimising the regeneration of tissues without provoking undesirable reactions in the host9. The aim of this study was to confirm the capacity for osteointegration of bioSiC ceramics in an animal model of a bone defect in rabbit, confirming its biocompatibility in vivo through a comparison of implants with bioactive coating and those without.Material and methods
Fabrication of bioinspired SiC
The process of making the bioinspired ceramics consists of two basic phases. The first is the process of pyrolysation, in which the natural precursor, dried and lyophilised is subject to a high temperature in inert conditions. The sample was heated at a controlled rate of 2°C/min up to 600-800°C. This temperature was maintained for 1 hour and then the sample was slowly cooled to room temperature. The resulting material is a carbonaceous structure/matrix which retains all the biostructural details of the vascular system of the original precursor.
The second phase consists of a process of infiltration, in which the carbon structure/matrix is covered with silicon powder and exposed to a high temperature in a controllable oven in vacuum conditions. A temperature of 1,550°C is reached, at a rate of 5-10°C/min, kept at the maximum temperature for 30 min, followed by a controlled cooling at 20°C/min, producing a piece of silicon carbide. For this study 36 cylinders of bioSiC derived from sapelli wood were made, of which 18 were coated with Si-HA (7.5 at.%) by means of pulsed laser deposit (PLD). The process of coating the Si-HA was carried out with an excimer laser (193 nm, 175 mJ and 10Hz) using a synthetic pellet of hydroxyapatite (HA) and silicon powder (7.5 at.% of Si). The deposit was performed in an atmosphere of water vapour (0.45 mbar) keeping the substrate at 460°C during the growth of the coating.

Animal model
To investigate the in vivo biocompatibility of the material an experimental model of a bone defect in a rabbit femur was designed. The cylinders of bioSiC were implanted in the lateral condyles of the femurs of 18 New Zealand rabbits. This was carried out by means of lateral longitudinal distal approach in the thighs of both extremities, and with a drill an orifice was made in the distal epiphysis of the femur in which was implanted, selected at random, a cylinder of bioSiC with no coating, or a cylinder of bioSiC with a bioactive coating of Si-HA. The animals were sacrificed using an intravenous administration of sodium phenobarbitol at 1, 4 and 12 weeks following the implant and samples were obtained, which were fixed in formol buffered at 10%. With this approach, 6 experimental groups were obtained (n=6), the first three with non-coated implants for the three periods of study, and the last three with coated implants. All the experiments were carried out in conformance with the Law 14/2007 and the Royal Decree 1201/2005, and following the directives of the UNE-EN rules 30993-3:1994 and ISO 10993-2:2006.

Histological analysis
The samples were processed for study using the techniques of embedding in methacrylate described by Donath10 following the stages of fixing, dehydration, infiltration, inclusion and polymerisation. Subsequently, longitudinal sections were made in the condyle of the femur approximately 30 µm thick, which were stained using the Lévai-Laczkó stain.
Once prepared, the samples were examined with a binocular microscope and the fraction of the total pore area occupied by bone, the quantity of neoformed bone in the periphery of the implant and the percentage of the surface of the implant in contact with bone were estimated11. The later calibration was carried out using the Microimage computer application.

Analysis using micro-CT
The formation of bone around the implant was also analysed using computerised microtomography (micro-CT). The samples were scanned using a SkyScan 1172 high resolution X-ray microtomograph (Bruker micro CT NV, Kontich, Belgium) with the intensity of the X-ray source at 60 kV and 167 µA. A nominal resolution of 7.9 µm was used and an Al filter of 1mm thickness used to obtain a restricted longitudinal wave interval. The rotation step was 0.2° with a total rotation of 360° and a frame averaging value (images per step) of three. The images obtained were reconstructed using the modified Feldkamp algorithm12 and analyses using the commercial application CTAnalyser (Bruker micro CT NV, Kontich, Belgium). For this, a volume of interest of 160 µm thickness from the surface of the implant was selected, in which was determined the bone volume fraction (BV/TV). In addition, the surface area of the intersection of the bone with the implant with reference to the total surface area of the implant (i.S/TS) was calculated.

Statistical analysis
The data gathered in this study were entered into a text database which was subsequently exported to the statistics software package SPSS 18.0 (IBM, Armonk, NY, USA) for their later statistical analysis. Then the descriptive analysis was made of the variables in the study. The descriptive statistics for the numerical variables were expressed as mean ± standard deviation. The comparative statistical study of all the numerical results obtained for the different groups of the study was carried out using the Mann-Whitney U test due to the fact that the variables did not exceed the normality criteria applied. The relationship between the results obtained by histology and micro-CT were studied using Pearson’s correlation.
The level of statistical significance was established at p<0.05 for all the variables analysed.

The in vitro biocompatibility of the bioSiC ceramics was demonstrated using a culture of the human osteoblast cell line MG-63 in an earlier study7. To investigate the biocompatibility in vivo of these bioceramics an experimental model in rabbit femur was used. The histological slides obtained for the samples after the implantation were examined using optical microscope and computerised microtomography.
Through the histological analysis it is possible to observe the growth of neoformed bone on the surface of the implants, with no signs of inflammation or appearance of fibrous tissue in the region of the samples. It is notable that the neoformed bone penetrates the pores of the implant. At higher magnifications of the microscope we can confirm this colonisation, as well as the bone’s contact with the implant, which is a key feature (Figure 1).
In the implants without coating, at the first week after implantation it is observed that 10.38% of the area of the pores have been colonised by new bone, a percentage which increases to 37.52% at 4 weeks (p=0.017). In the area selected in the periphery of the implant the neoformed bone after the first week occupied 21.25%, while at 4 weeks it was 31.30% (p=0.030). At 12 weeks no differences were observed in any variable with respect to the samples at 4 weeks (Figure 2).
In the analysis with micro-CT, the region of interest selected for the analysis of bone growth was 160 µm from the surface of the implant. After the first week from the positioning of the bioSiC implants, the BV/TV of the region analysed was 11.49%, which means that of the volume analysed around the implant, this percentage was occupied by bone. At 4 weeks, the BV/TV increased to 45.36% (p=0.030 vs 1st week). The value at 12 weeks showed no significant difference to that at 4 weeks. Similarly, the i.S/TS, which represents the percentage of the surface of the implant in contact with bone, increases up to the fourth week, after which its value appears to stabilise, leaving approximately half the surface of the implant in contact with neoformed bone. Again, no differences were found between the groups at 4 and 12 weeks (Figures 2 and 3).
The relationship between the results obtained using the two techniques was studied using Pearson’s correlation. The BV/TV calculated by micro-CT had a positive correlation with the percentage of neoformed bone in the periphery determined by histology (r=0.588, p<0.001); there was a similar relationship between the microtomographic intersection surface (i.S/TS) and the percentage of bone contact determined histologically (r=0.677, p<0.001) (Figure 4).

Nature offers a great variety of species with highly diverse levels of porosity, which means that we must select the original precursor most appropriate to the bone structure which we wish to replicate. Notable among their common characteristics is the disposition of having a hierarchical porous structure which is replicated in the bioinspired ceramics.
In the case of sapelli, we see a combination of microcanals with a diameter of between 80 and 100 µm, which have an impact on the promotion of vascularisation, and the transport of nutrients and waste products. Also observed are micropores of around 4 µm which participate in the formation of the capillaries7. Notable also is the presence of pores on the nanometric scale which play an interesting role in matters relating to molecular diffusion for nutrition and signalling. The interconnection of the pores provide a channel for cellular migration and allow the formation of blood vessels; and the roughness of the material contributes to increasing the surface area, favouring the adsorption of proteins and ion interchange3.
From the values obtained both through histological analysis and microstructural analysis with micro-CT it can be deduced that osteointegration is complete after 4 weeks. None of the parameters analysed by either of the techniques show significant differences between the samples at 4 and 12 weeks, either in the samples with the coating or those without. The samples in which a coating of Si-HA has been applied had no significant differences from the bioSiC samples without a coating for any of the variables analysed. From this fact it is deduced that the bioinspired SiC ceramics do not need the addition of a coating in order to be biocompatible, and that the coating does not improve its contact with bone.
The bioinspired ceramics of silicon carbide have, therefore, great potential as new materials for biomedical applications. The feasibility of producing strong lightweight ceramic devices with interconnected porosity suitable for use as bone substitutes has been demonstrated. The in vivo trials of bioSiC implants indicate that the pores of this material are colonised by bone tissue and support its mineralisation. They also confirm that they have a good apposition for neoformed bone and good osteointegration in a relatively short time after implantation.

1. Salgado AJ, Coutinho OP, Reis RL. Bone tissue engineering: State of the art and future trends. Macromol Biosci 2004;4:743-65.
2. Vicario Espinosa C. Los aloinjertos óseos en Cirugía Ortopédica y Traumatología (1). Patología Aparato Locomot 2004;2:214-32.
3. Ratner BD, Hoffman AS, Schoen FJ, Lemons JE (eds). Biomaterials Science: Introduction to Materials in Medicine, 2nd Edition. London: Elsevier Academic Press; 2004.
4. Martínez-Fernández J, Varela-Feria FM, Singh M. High temperature compressive mechanical behavior of biomorphic silicon carbide ceramics. Scr Mater 2000;43:813-8.
5. De Arellano-López AR, Martínez-Fernández J, González P, Domínguez C, Fernández-Quero V, Singh M. Biomorphic SiC: A new engineering ceramic material. Int J Appl Ceram Technol 2004;1:56-67.
6. Kardashev BK, Burenkov YuA, Smirnov BI, de Arellano-López AR, Martínez-Fernández J, Varela-Feria FM. Elasticity and inelasticity of biomorphic silicon carbide ceramics. Phys Solid State 2004;46:1873-7.
7. González P, Borrajo JP, Serra J, Chiussi S, León B, Martínez-Fernández J, et al. A new generation of bio-derived ceramic materials for medical applications. J Biomed Mater Res A 2009;88:807-13.
8. An YH. Mechanical properties of bone. In: An YH, Draughn RA (eds.). Mechanical testing of bone and the bone-implant interface. Boca Raton, Florida, (USA): CRC Press; 2000.
9. Williams D. Revisiting the definition of biocompatibility. Med Device Technol 2003;14:10-3.
10. Donath K. Preparation of histologic sections by the cutting-grinding technique for hard tissue and other material not suitable to be sectioned by routine methods. Norderstedt: Institute for Pathologie, University of Hamburg; Exakt-Kulzer-Publication; 1995.
11. Tonino AJ, Thèrin M, Doyle C. Hydroxyapatite-coated femoral stems: Histology and histomorphometry around five components retrieved at post mortem. J Bone Joint Surg Br 1999;81:148-54.
12. Feldkamp LA, Davis LC, Kress JW. Practical cone-beam algorithm. J Opt Soc Am A 1984;1:612-9.