- Email: [email protected]
Advances in the ultrastructural study of the implant–bone interface by backscattered electron imaging
Available online at www.sciencedirect.com
Micron 39 (2008) 1363–1370 www.elsevier.com/locate/micron
Advances in the ultrastructural study of the implant–bone interface by backscattered electron imaging J. Wierzchos a,*, T. Falcioni b, A. Kiciak c, J. Wolin´ski d, R. Koczorowski e, P. Chomicki f, M. Porembska f, C. Ascaso g a
Servei de Microscopia Electronica, Universitat de Lleida, c/Rovira Roure 44, 25198 Lleida, Spain b Departamento Cientı´fico, Ilerimplant S.L., c/Enginyer Mı´es, 705 A Lleida, Spain c Department of General and Gastroenterological Surgery, Orlowski Hospital, MCPE, 231 Czerniakowska Street, 00-416 Warsaw, Poland d Kielanowski’s Institute, Polish Academy of Sciences, 3 Instytucka Street, 05-110 Jabłonna, Poland e Department of Prosthetic Dentistry, University of Medical Sciences, Collegium Stomatologicum, 70 Bukowska Street, 60-812 Poznan´, Poland f Department of Cranio-Maxillo-Facial Surgery, Oral Surgery and Implantology, 4 Lindley’a Street, 02-005 Warsaw, Poland g Instituto de Recursos Naturales, CCMA, CSIC, c/Serrano 115 dpdo, 28006 Madrid, Spain Received 10 July 2007; received in revised form 16 January 2008; accepted 17 January 2008
Abstract The biocompatibility of titanium implants in bone depends on the response shown by cells in contact with the implant surface. Several developments have been targeted at achieving successful implant treatment. The aim of this study was to develop a novel preparation procedure to evaluate the bone cell response produced at the bone–implant interface using the technique scanning electron microscopy with backscattered electron imaging (SEM-BSE). Dental prostheses with an SLA-modified or TOP-modified surface were implanted in a toothless part of the mandibula in female pigs. The animals were sacrificed 12 weeks after surgery, at which time block specimens containing the implants were obtained. These specimens were then processed for SEM-BSE by optimizing a protocol involving chemical fixation and heavy metal staining. In addition, element distribution maps for the implant–bone tissue interface were obtained using a microanalytical system based on energy-dispersive X-ray spectrometry (EDS). This novel visualisation approach enabled a comprehensive study of the extracellular matrix and cell components of the host tissues neoformed around the implant. SEM-BSE images also provided ultrastructural details of the bone cells. This technique appears to be an effective and very promising tool for detailed studies on the implant–bone tissue interface and the host response to the bone incorporation process. # 2008 Elsevier Ltd. All rights reserved. Keywords: Animal study; Bone cells; Bone formation; Implants; Interface; Osseointegration; SEM-BSE
1. Introduction The biocompatibility of titanium implants in bone depends on the interfacial response of cells in contact with the implant surface (Huang et al., 2004). The final biological response of bone to an implanted biomaterial depends on the implant’s bulk properties, its design and specific surface characteristics, which will affect the tissue repair process and the long-term behaviour of the biomaterial–tissue interface in the recipient patient (Savarino et al., 2003). The surface properties of prostheses play an important role in how they interact with adjacent living
* Corresponding author. Tel.: +34 973702417; fax: +34 973702426. E-mail address: [email protected] (J. Wierzchos). 0968-4328/$ – see front matter # 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.micron.2008.01.022
tissue (Li et al., 2006). Several modifications have been made to implant surfaces to improve their incorporation in recipient bone tissue. In vitro and in vivo studies have revealed that the cell biology and shear strength of the bone–implant interface are influenced by the microtopography of the implant surface (Lincks et al., 1998; Deligianni et al., 2001). It is also widely known that bone calcification is preceded by the formation of an extracellular organic matrix, primarily secreted by bone cells. Bone-forming cells, active osteoblasts and osteocytes, control the synthesis of this extracellular matrix, and regulate the exchange between ions present in the bulk extracellular fluids and those present in calcified bone. Thus, a detailed in situ study and the precise visualization of all the components of the interface between the implant surface and newly formed bone are needed to assess the bone integration process.
1364
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
Several techniques have been used to assess the implant– bone tissue interface. The traditional method of histological staining followed by examination under the light microscope (LM) has enabled observation of the cell response produced during bone regeneration. Although LM has contributed substantial information, its low spatial resolution determines that no information is provided at the ultrastructural level. Transmission electron microscopy (TEM), on the other hand, has been successfully used to describe the cellular components of newly formed bone (e.g. Meyer et al., 2004). However, the technique has an inherent limitation: only soft tissue mechanically separated from the implant–bone interface that has been processed to prepare ultra-thin sections can be examined. This along with the small size of the observed area (1–2 mm2) in most cases adds to its limitations. Among the most promising tools as a complementary technique to both LM and TEM, is scanning electron microscopy (SEM) operating in backscattered electron (BSE) emission mode. The use of the SEM-BSE technique to examine the implant–bone interface was first reported by Jasty et al. (1989), Bloebaum et al. (1990) and Skedros et al. (1993a). The method was mostly used for descriptive studies performed on calcified tissue (Skedros et al., 1993a,b; Boyde and Jones, 1996; Roschger et al., 1997). Other authors have reported the use of SEM-BSE to examine fracture healing (Egger et al., 1993; Kim et al., 1992) and the bone response to the implant surface (Sul et al., 2005; Ottani et al., 2002). In addition, SEM-BSE imaging has unveiled some of the ultrastructural features of the different calcified tissues involved in the process of bone healing in studies conducted by Franch et al. (1998) and Manzanares Ce´spedes et al. (2004). These authors argue this technique to be one of the best methods for the morphological characterization of calcified tissues in studies designed to evaluate the fracture repair process. Despite obvious advantages of the SEM-BSE technique, so far it has only been used to describe and characterize calcified bone patterns and not for histological purposes. The principal objective of the present study was to develop a novel protocol for preparing specimens of dental implants embedded in newly formed bone for SEM-BSE observation, for use in studies designed to assess the implant–bone interface. The idea was to achieve sufficient resolution to explore the different inorganic and biological processes of tissue healing that occur at this interface. The approach developed is also useful for the ultrastructural characterization of bone cells and their components involved in the bone integration process. We also used the energy-dispersive X-ray spectrometry (EDS) microanalytical system to prepare element distribution maps of the implant–bone tissue interface.
sandblasted, acid-etched (SLA) surface (www.ilerimplant.com); and the Nobel TiUniteTM (Nobel Biocare AB, Go¨teborg, Sweden), also composed of titanium with a novel titanium porous oxide (TPO) coating (Hall and Lausmaa, 2000). 2.2. Animals and surgical technique Eight female Polish Landrace pigs with an initial body weight (b.w.) ca. 250 kg were obtained from a commercial piggery and fed twice a day with a feed concentrate as rations (2% b.w.) and water ad libitum. The animal were individually caged and kept under constant conditions of light (12/12 h light/ dark cycle) at room temperature, Azaperone (4 mg/kg b.w., Stressnil, Janssen Belgium) was injected intramuscularly as premedication and anaesthesia achieved with pentobarbiturate at the routinely used dose for second stage surgical anaesthesia. Antibiotic prophylaxis consisted of two injections (15 mg/kg b.w., amoxycillin, Clamoxyl L.A., Pfizer, UK) given 1 h before implant and continued for a further 48 h. Four animals were implanted with an SLA-modified surface implant and a further four animals received a TOP-modified surface implant. Following a gingival incision, the bone was gradually penetrated using the lancet and round pilot drills to the required depth, as measured using a depth gauge. Next, a series of consecutive holes were drilled to cover a diameter appropriate for each implant and up to the required depth mark. The implants were then inserted in the prepared site until the end of the most coronal aspect of the fixture’s microthread. A contra-angle hand-piece was introduced for motorized installation with a torque strength not exceeding 40 N cm and maximum speed 15–20 rpm. If the implant was not completely inserted using the contra-angle protocol, insertion was completed using a torque wrench (maximum torque strength 40 N cm) and its screwtool. There was no need to use the screw tap in any case. The implant abutments were then sealed using the cover screws at the top of the implant holder cap. Finally, the gingiva was sutured using Maxon 3–0 absorbable suture thread (Syneture, Tyco Healthcare Group LP, USA). Surgery was performed by experienced implantologists using the techniques commonly used in humans. Treatments and experiments were conducted according to EU guidelines on the welfare of experimental animals. The study protocol was approved by the local ethics committee (application no. 40/ 2005, decision no. 41/2005). At week 12 post-implantation, following sedation with ketamine (10 mg/kg, i.m.) the animals were sacrificed by pentobarbiturate (Thiopental, Biochemie GMbH, Austria) overdose. 2.3. Preparing samples for SEM-BSE
2. Materials and methods 2.1. Oral implants Two types of dental implants available in the market were used: the ‘‘Phoenix’’ implant (Ilerimplant1 S.L.-GMI S.L., Lleida, Spain), a solid screw (outer diameter 4 mm, implanted segment 10 mm) made of c.p. titanium (grade IV) with a
Following animal sacrifice, specimens comprising the dental implants and adjacent host tissue were excised from the mandibular bone as cubes (15 mm 15 mm 15 mm) and immediately transferred to 2.5% glutaraldehyde in 0.1 M phosphate buffer (PB). The samples were chemically fixed in this solution for 24 h at 4 8C. Once fixed, the samples were washed in 0.1 M PB 3 30 min and post-fixed in 1% osmium
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
tetroxide in H2O for 24 h at 4 8C. The specimens were again washed in 0.1 M PB (3 30 min). Next, the samples were dehydrated in a graded ethanol series of 30% for 3 h followed by 50% for 3 h. During dehydration with the next ethanol dilution, the samples were contrasted with saturated 70% ethanol uranyl acetate overnight at 4 8C. This was followed by immersion in 96% (2 3 h) and finally 100% (3 3 h) ethanol. Next, the samples were gradually infiltrated with LRWhite (The London Resin Co. Ltd., Hampshire, UK) embedding medium, first with a 1:1 mixture of LRW in 100% ethanol (3 days at 4 8C) and finally with pure LRW resin (3 3 days at 4 8C). This step was followed by polymerisation in an oxygen-free atmosphere (24 h, 60 8C). After this, the polymerised specimen blocks were visualized by X-ray microradiography to determine the orientation of the implants within the tissue. This helps achieve perfect cross-sectioning through the longitudinal axis of the implant. Next, the blocks were cut using a low speed diamond saw and fine polished using grinding papers (from nr. 300 to 1200) and silicone carbide abrasive sheets with grain diameters of 15, 9 and 3 mm. Final polishing was performed using liquid diamond polishing compound on napped cloth with diamond particle sizes of 1 and 0.25 mm in an oil-based lubricant fluid. The surfaces of the polished cross-sections were post-stained with saturated uranyl acetate (in H2O) for 30 min and with lead citrate (Reynolds, 1963) for 12 min. After washing in distilled water and airdrying, the polished block surfaces were coated with evaporated carbon. 2.4. SEM-BSE microscopy and EDS microanalysis Prepared samples were examined in a scanning electron microscope (DSM940 Zeiss) equipped with a solid-state, four diode BSE detector plus an auxiliary EDS microanalytical system (Link ISIS Oxford). The EDS method involved qualitative and quantitative microanalysis, including element distribution mapping. In the distribution maps provided, relative concentrations are indicated by a colour scale: darkblue represents a concentration of absolute zero and white denotes 100% absolute concentration of the respective purecomponent spectrum (Kotula et al., 2003). The colour concentration scales (0–100%) for each element are provided in the corresponding images. The microscopy and/or microanalytical operating conditions were as follows: 08 tilt angle, 358 take-off angle, 15 kV acceleration potential, 6 or 25 mm working distance and 1–5 nA specimen current. 3. Results The results presented indicate the possible application of the SEM-BSE technique to examine the implant–new bone tissue interface of the specimens prepared using the new procedure. Fig. 1 shows different aspects of the SLA-modified titanium implant–bone interface observed at 12 weeks post-implant. The low magnification view of this interface (Fig. 1a) shows the titanium implant surrounded by newly formed lamellar bone revealing a Haversian, or osteonal bone, pattern. Details of the
1365
Haversian canal, appearing close to the implant surface, are shown in transverse view in Fig. 1b. Note that the wall of this osteon is overlain by a layer of deposited bone tissue and osteoblast cells. According to Skedros et al. (2007), the irregular shape and relatively large diameter of this vascular channel suggests that this is the secondary osteon. The darker contrast of the lamella layer in the SEM-BSE image indicates the lower mean atomic number of the target, probably attributable to the lower content of apatitelike crystals within this structure. In the centre of this canal, a blood vessel containing erythrocytes could be distinguished. Details of the proper integration of the SLA-modified surface implant with the bone are shown in Fig. 1c. A relative abundance of osteocytes was observed in this mineralized mature bone. In some zones of the implant–bone interface, the implant surface was overlain by an extracellular matrix containing bone (osteogenic) cells (Fig. 1d and e). The calcium EDS distribution map (Fig. 1f) clearly demonstrates differences in mineralisation rates (reflected by Ca concentrations) between mineralized bone and immature bone (osteoid), as well as small amounts of calcium in the extracellular matrix. Fig. 2 provides information on the osseointegration process produced in the implants coated with highly porous titanium oxide (TPO). The external micromorphology of this layer (before implant) can be seen in Fig. 2a. Quantitative EDS analysis of this layer revealed a titanium oxide (TiO2) concentration (wt%) of 83.78% (n = 15, S.D. = 1.23). However, this layer was also found to contain a small quantity of phosphorus (the P concentration represented as P2O5 was 16.03%; n = 15, S.D. = 0.37). Our SEM-BSE images (Fig. 2b, c and e) revealed the mineralized bone to be tightly adhered to the TPO layer after 12 weeks of implant. Many of this layer’s conical- and bottle-shaped pores were filled with a matrix composed of calcium, phosphorus and oxygen (EDS map in Fig. 2d). Moreover, the EDS elemental distribution maps (Fig. 2d) distinctly show how the osseous tissue composed of calcium hydroxyapatite adheres to the TPOmodified surface, as well as revealing the appearance of the phosphorous and oxygen elements in the TPO layer (dotted line in Fig. 2d). Fig. 2f shows the occurrence of osteogenic cells close to the implant surface. The SEM-BSE images in Fig. 3 indicate the potential of the proposed strategy to detect and characterize the bone cells present at the implant–neoformed bone interface. Even a low magnification image can provide interesting information on the contents of Haversian canal (Fig. 3a). The detailed view of the centre of the Haversian canal (Fig. 3b) shows osteogenic cells embedded in collagen fibrils. Additionally, the resolution power of this technique can render ultrastructural details of osteocyte-containing lacunae in lamellar bone. Osteocytes are star-shaped cells, and are the most abundant cell type found in bone. The most conspicuous feature of the osteocyte that appears in Fig. 3c is the round nucleus with nucleolus and endoplasmic reticulum. The next image (Fig. 3d) shows mature bone covered with an extracellular matrix of noncalcified bone. Osteoblasts were also frequently observed overlying mature bone (Fig. 3e). In close proximity to these osteoblasts, osteogenic (bone progenitor) cells could also be seen in the
1366
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
Fig. 1. SEM-BSE and EDS cross-section images of the SLA-modified (Phoenix) titanium implant. (a) General view of a titanium (Ti) implant and surrounding bone (B). (b) Detailed view of the area indicated in (a): Ti, implant; B, bone; open arrow, Haversian canal; black arrow, lamella; white arrow, osteoblasts; white arrowhead, erythrocytes. (c) Rough Ti implant surface closely attached to mature mineralized bone (B); white arrows, osteocytes. (d) Osteogenic cells (arrows) and erythrocytes (heads) in the extracellular matrix adjacent to the Ti implant surface. (e) Mineralized bone (B), osteoid (Ot, outlined by a dotted line) and extracellular matrix containing osteogenic cells adjacent to the Ti implant surface. The space (#) between the coating and the Ti implant is an artefact of the tissue preparation process. (f) EDS distribution map showing the calcium (Ca) concentration of the area shown in (e).
extracellular matrix. Odontoblasts (Fig. 3f) were also observed in zones of osseous tissue further from the implant. 4. Discussion The placement of an implant in the alveolar process elicits a sequence of healing events including necrosis and subsequent
resorption of traumatized bone around the titanium fixture concomitant with new bone formation. While the implant initially displays mechanical stability due to contact and friction between the implant surface and the severed bone, the long-term maintenance of implant stability calls for biological attachment between the foreign body and the surrounding tissue. The precision and reliability of a histomorphometric
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
1367
Fig. 2. SEM and EDS cross-section images of the TPO-modified (TiUniteTM) titanium implant. (a) Image obtained by SEM in secondary electron mode showing details of the TPO-modified surface of the TiUniteTM implant. (b) and (c) detailed SEM-BSE micrographs of the interface zone between the Ti implant, TPO-modified implant surface and mineralized bone (B). Note how the bone matrix infills the conical-shaped (open arrow in (b) and bottle-shaped (black arrow in (c) pores of the TPO layer. (d) EDS distribution maps showing Ti, P, Ca and O concentrations, respectively, of the area shown in (c); arrows indicate calcium within the pores of TPO layer. (e) Detailed SEM-BSE view of a bottle-shape pore (boxed area in (c) filled with a Ca–P–O-rich matrix. (f) SEM-BSE micrograph showing a Ti implant and extracellular matrix containing osteogenic cells (arrows) adjacent to the TPO surface.
study of the newly formed bone depends upon the correct identification and ultrastructural characterization of all possible cellular components that play a role in the osseointegration processes. The LM lacks the resolution power required for a detailed structural analysis. The use of electron microscopy techniques could help characterize the morphological changes that occur during osseointegration.
The main electron beam in SEM causes the emission of BSE from the specimen. BSE may be used to detect contrast between areas of different chemical composition. These can be observed especially when the average atomic numbers of the components of the regions vary. Since the intensity of the BSE signal depends on the mean atomic number of the sample (Joy, 1991), the SEMBSE technique not only serves to distinguish inorganic features,
1368
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
Fig. 3. SEM-BSE images of bone cells close to the implant–bone interface. (a) General view of a longitudinal section of a Haversian canal in lamellar bone (B); black arrows, osteocytes; white arrows, osteoblast cells. (b) Detailed view of the area indicated in (a) showing osteogenic cells and collagen fibrils (F). (c) Lacuna occupied by an osteocyte cell in lamellar bone (B); n, nucleus; black arrow, nucleolus; white arrows, endoplasmic reticulum. (d) Mature bone (B) overlain by an extracellular matrix of noncalcified bone (E) with adjacent osteoblast cells; n, nuclei; black arrow, nucleolus. (e) Mature bone (B) overlain by a layer of osteoblast cells (open arrow); black arrows, nucleoli of osteoblast nuclei; white arrows, osteogenic cells. (f) Detailed view of odontoblasts (white arrows, nuclei) and erythrocytes (black arrow) within a blood capillary.
but also offers the interesting possibility of identifying heavy metal-stained ultrastructural cell components and their micromorphological details. The new approach presented in this study was inspired by the successful application of the SEM-BSE technique and heavy metal staining preparation procedure to the study of the microorganism–rock interface. This new research tool was reported first time by Wierzchos and Ascaso (1994). The
present report describes a first attempt at applying this method to characterize the implant–bone tissue interface. SEM-BSE imaging has the advantage of a wide magnification range determining that pieces of bone harbouring implants as small as several centimetres can be scanned. When the site of interest has been identified, it can be magnified and imaged at high-resolution close to that of TEM. Moreover, microanaly-
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
tical determinations in the form of both qualitative and quantitative measurements and element distribution maps can be simultaneously generated for the area of interest. For the stability of a load-bearing implant, the formation of new bone and a mineralized matrix near the implant are key processes. The present study revealed that highly mineralized bone became closely attached to both the SLA- and TPOmodified implant surfaces. Moreover, it was observed (Fig. 2b–e) that the small pores in the TPO-modified implant surface were filled with a Ca–P–O-rich matrix, suggesting the penetration of newly formed bone tissue. This could confer these implants the desired mechanical stability. The presence of small amounts of calcium was also observed in the bone-specific extracellular matrix produced by bone cells concomitant with apatite crystal formation. These findings demonstrating calcification of the bone tissue at the implant–bone interface are in agreement with previous reports, e.g. Simmons et al. (1999). Our ultrastructural data indicate that cells with the typical appearance of welldifferentiated bone cells occur in very close proximity (micrometers) to the implant surface. This observation is consistent with those described by Lavos-Valereto et al. (2001), Simmons et al. (1999) and Meyer et al. (2004) who also noted intimate contact between bone cells and titanium implants on SEM examination in both early and late post-implant stages. 5. Conclusions SEM-BSE visualization of heavy metal-stained bone tissue is very useful for assessing the behaviour of an implant at the bone interface. SEM-BSE images of finely polished crosssections of specimens taken from the contact zone between a titanium implant and surrounding neoformed bone can provide relevant information on the presence and ultrastructural details of bone cells occurring at the implant–bone interface. Moreover, by visualizing the interactions that occur at this interface between cells and their components we will increase our understanding of the osseointegration process. Acknowledgements This research was supported by a formal agreement between the University of Lleida and the firm Ilerimplant S.L., in cooperation with the Kielanowski Institute, the Polish Academy of Sciences and by the Departamento de Universidades, Investigacio´n y Sociedad de la Informacio´n (DURSI) de la Generalidad de Catalun˜a. The authors thank A. Burton for help with preparing the English version of the manuscript and X. Calomarde for help with the sample preparation and technical procedures. The authors are also indebted to J. Esquerda and O. Trabal for their help in interpreting some of the images and J. Qui for X-ray micro-radiography facility. References Bloebaum, R.D., Bachus, K.N., Boyce, T.M., 1990. Backscattered electron imaging: the role in calcified tissue and implant analysis. J. Biomater. Appl. 5, 56–85.
1369
Boyde, A., Jones, S.J., 1996. Scanning electron microscopy of bone: instrument, specimens and tissues. Microsc. Res. Tech. 33, 92–120. Deligianni, D.D., Katsala, N., Ladas, S., Sotiropoulou, D., Amedee, J., Missirlis, Y.F., 2001. Effect of surface roughness of the titanium alloy Ti–6Al– 4V on human bone marrow cell response and on protein adsorption. Biomaterials 22, 1241–1251. Egger, E.L., Gottsauner-Wolf, F., Palmer, J., Aro, H.T., Chao, E.Y.S., 1993. Effects of axial dynamization on bone healing. J. Trauma 34, 185–192. Franch, J., Garcı´a, F., Camo´n, J., Manzanares, M.C., 1998. Backscattered electron imaging of the calcified tissues present in bone healing. Vet. Comp. Orthop. Traumatol. 11, 105–111. Hall, J., Lausmaa, J., 2000. Properties of a new porous oxide surface on titanium implants. Appl. Osseointegr. Res. 1, 5–8. Huang, H.H., Ho, C.T., Lee, T.H., Lee, T.L., Liao, K.K., Chen, F.L., 2004. Effect of surface roughness of ground titanium on initial cell adhesion. Biomol. Eng. 21, 93–97. Jasty, M., Bragdon, C.R., Schtzer, S., Rubash, H., Haire, T., Harris, W.H., 1989. Bone ingrowth into porous coated canine total hip replacements. Quantification by backscattered scanning electron microscopy and image analysis. Scan. Microsc. 3, 1051–1057. Joy, D.C., 1991. An introduction to Monte Carlo simulation. Scan. Microsc. 5, 329–337. Kim, W., Caja, V.L., Larsson, S.A.B., Chao, E.Y.S., 1992. Advanced histomorphometric technique on the analysis of early fracture repair under external fixation. Int. J. Orthop. Trauma Suppl. 3, 5–10. Kotula, P.G., Keenan, M.R., Michael, J.R., 2003. Automated analysis of SEM X-ray spectral images: a powerful new microanalysis tool. Microsc. Microanal. 9, 1–17. Lavos-Valereto, I.C., Wolynec, S., Deboni, M.C., Koing Jr., B., 2001. In vitro and in vivo biocompatibility testing of Ti–6AI–7Nb alloy with and without plasma-sprayed hydroxyapatite coating. J. Biomed. Mater. Res. 58, 727–733. Li, C.Y., Gao, S.Y., Terashita, T., Shimokawa, T., Kawahara, H., Matsuda, S., Kobayashi, N., 2006. In vitro assays for adhesion and migration of osteoblastic cells (Saos-2) on titanium surfaces. Cell Tissue Res. 324, 369–375. Lincks, J., Boyan, B.D., Blanchard, C.R., Lohmann, C.H., Liu, Y., Cochran, D.L., Dean, D.D., Schwartz, Z., 1998. Response of MG63 osteoblast like cells to titanium and titanium alloy is dependent on surface roughness and composition. Biomaterials 19, 2219– 2232. Manzanares Ce´spedes, M.C., Franch Serracanta, J., Carvalho Lobato, P., Montesdeoca Zaeza, J.M., Garcı´a Molina, J.A., Greco Machado, Y., Belmonte Caldero´n, A.M., Delafuente Baigorri, M.P., Serra Renom, I., 2004. The role of calcified tissues in the different stages of implant osteointegration. Dent. Med. Probl. 41, 179–191. Meyer, U., Joos, U., Mythili, J., Stamm, T., Hohoff, A., Fillies, T., Stratmann, U., Wiesmann, H.P., 2004. Ultrastructural characterization of the implant/ bone interface of immediately loaded dental implants. Biomaterials 25, 1959–1967. Ottani, V., Raspanti, M., Martini, D., Tretola, G., Ruggeri Jr., Franchi, M., Giuliani Piccari, G., Ruggeri, A., 2002. Electromagnetic stimulation of bone growth using backscattered electron imaging. Micron 33, 121–125. Reynolds, S., 1963. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell Biol. 17, 200–211. Roschger, P., Fratzl, P., Klaushofer, K., Rodan, G., 1997. Mineralization of cancellous bone after alendronate and sodium fluoride treatment: a quantitative backscattered electron imaging study on minipig ribs. Bone 20, 393–397. Savarino, L., Fini, M., Ciapetti, G., Cenni, E., Granchi, D., Baldini, N., Greco, M., Rizzi, G., Giardino, R., Giunti, A., 2003. Biologic effects of surface roughness and fluorhydroxyapatite coating on osteointegration in external fixation systems: an in vivo experimental study. J. Biomed. Mater. Res. Part A 66, 652–661. Simmons, C.A., Valiquette, N., Pilliar, R.M., 1999. Osseointegration of sintered porous-surfaced and plasma-spray coated implants: an animal model study
1370
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
of early post-implantation healing response and mechanical stability. J. Biomed. Mater. Res. 47, 127–138. Skedros, J.G., Bloebaum, R.D., Bachus, K.N., Boyce, T.M., 1993a. The meaning of greylevels in backscattered electron images of bone. J. Biomed. Mater. Res. 27, 47–56. Skedros, J.G., Bloebaum, R.D., Bachus, K.N., Boyce, T.M., Constantz, B., 1993b. Influence of mineral content and composition on graylevels in backscattered electron images of bone. J. Biomed. Mater. Res. 27, 57–64.
Skedros, J.G., Sorenson, S.M., Jenson, N.H., 2007. Are distributions of secondary osteon variants useful for interpreting load history in mammalian bones? Cell Tissues Organs 185, 285–3007. Sul, Y.T., Johansson, C., Byon, E., Alberktsson, T., 2005. The bone response of oxidized bioactive and non-bioactive titanium implants. Biomaterials 26, 6720–6730. Wierzchos, J., Ascaso, C., 1994. Application of back-scattered electron imaging to the study of the lichen-rock interface. J. Microsc. 175, 54–59.
Micron 39 (2008) 1363–1370 www.elsevier.com/locate/micron
Advances in the ultrastructural study of the implant–bone interface by backscattered electron imaging J. Wierzchos a,*, T. Falcioni b, A. Kiciak c, J. Wolin´ski d, R. Koczorowski e, P. Chomicki f, M. Porembska f, C. Ascaso g a
Servei de Microscopia Electronica, Universitat de Lleida, c/Rovira Roure 44, 25198 Lleida, Spain b Departamento Cientı´fico, Ilerimplant S.L., c/Enginyer Mı´es, 705 A Lleida, Spain c Department of General and Gastroenterological Surgery, Orlowski Hospital, MCPE, 231 Czerniakowska Street, 00-416 Warsaw, Poland d Kielanowski’s Institute, Polish Academy of Sciences, 3 Instytucka Street, 05-110 Jabłonna, Poland e Department of Prosthetic Dentistry, University of Medical Sciences, Collegium Stomatologicum, 70 Bukowska Street, 60-812 Poznan´, Poland f Department of Cranio-Maxillo-Facial Surgery, Oral Surgery and Implantology, 4 Lindley’a Street, 02-005 Warsaw, Poland g Instituto de Recursos Naturales, CCMA, CSIC, c/Serrano 115 dpdo, 28006 Madrid, Spain Received 10 July 2007; received in revised form 16 January 2008; accepted 17 January 2008
Abstract The biocompatibility of titanium implants in bone depends on the response shown by cells in contact with the implant surface. Several developments have been targeted at achieving successful implant treatment. The aim of this study was to develop a novel preparation procedure to evaluate the bone cell response produced at the bone–implant interface using the technique scanning electron microscopy with backscattered electron imaging (SEM-BSE). Dental prostheses with an SLA-modified or TOP-modified surface were implanted in a toothless part of the mandibula in female pigs. The animals were sacrificed 12 weeks after surgery, at which time block specimens containing the implants were obtained. These specimens were then processed for SEM-BSE by optimizing a protocol involving chemical fixation and heavy metal staining. In addition, element distribution maps for the implant–bone tissue interface were obtained using a microanalytical system based on energy-dispersive X-ray spectrometry (EDS). This novel visualisation approach enabled a comprehensive study of the extracellular matrix and cell components of the host tissues neoformed around the implant. SEM-BSE images also provided ultrastructural details of the bone cells. This technique appears to be an effective and very promising tool for detailed studies on the implant–bone tissue interface and the host response to the bone incorporation process. # 2008 Elsevier Ltd. All rights reserved. Keywords: Animal study; Bone cells; Bone formation; Implants; Interface; Osseointegration; SEM-BSE
1. Introduction The biocompatibility of titanium implants in bone depends on the interfacial response of cells in contact with the implant surface (Huang et al., 2004). The final biological response of bone to an implanted biomaterial depends on the implant’s bulk properties, its design and specific surface characteristics, which will affect the tissue repair process and the long-term behaviour of the biomaterial–tissue interface in the recipient patient (Savarino et al., 2003). The surface properties of prostheses play an important role in how they interact with adjacent living
* Corresponding author. Tel.: +34 973702417; fax: +34 973702426. E-mail address: [email protected] (J. Wierzchos). 0968-4328/$ – see front matter # 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.micron.2008.01.022
tissue (Li et al., 2006). Several modifications have been made to implant surfaces to improve their incorporation in recipient bone tissue. In vitro and in vivo studies have revealed that the cell biology and shear strength of the bone–implant interface are influenced by the microtopography of the implant surface (Lincks et al., 1998; Deligianni et al., 2001). It is also widely known that bone calcification is preceded by the formation of an extracellular organic matrix, primarily secreted by bone cells. Bone-forming cells, active osteoblasts and osteocytes, control the synthesis of this extracellular matrix, and regulate the exchange between ions present in the bulk extracellular fluids and those present in calcified bone. Thus, a detailed in situ study and the precise visualization of all the components of the interface between the implant surface and newly formed bone are needed to assess the bone integration process.
1364
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
Several techniques have been used to assess the implant– bone tissue interface. The traditional method of histological staining followed by examination under the light microscope (LM) has enabled observation of the cell response produced during bone regeneration. Although LM has contributed substantial information, its low spatial resolution determines that no information is provided at the ultrastructural level. Transmission electron microscopy (TEM), on the other hand, has been successfully used to describe the cellular components of newly formed bone (e.g. Meyer et al., 2004). However, the technique has an inherent limitation: only soft tissue mechanically separated from the implant–bone interface that has been processed to prepare ultra-thin sections can be examined. This along with the small size of the observed area (1–2 mm2) in most cases adds to its limitations. Among the most promising tools as a complementary technique to both LM and TEM, is scanning electron microscopy (SEM) operating in backscattered electron (BSE) emission mode. The use of the SEM-BSE technique to examine the implant–bone interface was first reported by Jasty et al. (1989), Bloebaum et al. (1990) and Skedros et al. (1993a). The method was mostly used for descriptive studies performed on calcified tissue (Skedros et al., 1993a,b; Boyde and Jones, 1996; Roschger et al., 1997). Other authors have reported the use of SEM-BSE to examine fracture healing (Egger et al., 1993; Kim et al., 1992) and the bone response to the implant surface (Sul et al., 2005; Ottani et al., 2002). In addition, SEM-BSE imaging has unveiled some of the ultrastructural features of the different calcified tissues involved in the process of bone healing in studies conducted by Franch et al. (1998) and Manzanares Ce´spedes et al. (2004). These authors argue this technique to be one of the best methods for the morphological characterization of calcified tissues in studies designed to evaluate the fracture repair process. Despite obvious advantages of the SEM-BSE technique, so far it has only been used to describe and characterize calcified bone patterns and not for histological purposes. The principal objective of the present study was to develop a novel protocol for preparing specimens of dental implants embedded in newly formed bone for SEM-BSE observation, for use in studies designed to assess the implant–bone interface. The idea was to achieve sufficient resolution to explore the different inorganic and biological processes of tissue healing that occur at this interface. The approach developed is also useful for the ultrastructural characterization of bone cells and their components involved in the bone integration process. We also used the energy-dispersive X-ray spectrometry (EDS) microanalytical system to prepare element distribution maps of the implant–bone tissue interface.
sandblasted, acid-etched (SLA) surface (www.ilerimplant.com); and the Nobel TiUniteTM (Nobel Biocare AB, Go¨teborg, Sweden), also composed of titanium with a novel titanium porous oxide (TPO) coating (Hall and Lausmaa, 2000). 2.2. Animals and surgical technique Eight female Polish Landrace pigs with an initial body weight (b.w.) ca. 250 kg were obtained from a commercial piggery and fed twice a day with a feed concentrate as rations (2% b.w.) and water ad libitum. The animal were individually caged and kept under constant conditions of light (12/12 h light/ dark cycle) at room temperature, Azaperone (4 mg/kg b.w., Stressnil, Janssen Belgium) was injected intramuscularly as premedication and anaesthesia achieved with pentobarbiturate at the routinely used dose for second stage surgical anaesthesia. Antibiotic prophylaxis consisted of two injections (15 mg/kg b.w., amoxycillin, Clamoxyl L.A., Pfizer, UK) given 1 h before implant and continued for a further 48 h. Four animals were implanted with an SLA-modified surface implant and a further four animals received a TOP-modified surface implant. Following a gingival incision, the bone was gradually penetrated using the lancet and round pilot drills to the required depth, as measured using a depth gauge. Next, a series of consecutive holes were drilled to cover a diameter appropriate for each implant and up to the required depth mark. The implants were then inserted in the prepared site until the end of the most coronal aspect of the fixture’s microthread. A contra-angle hand-piece was introduced for motorized installation with a torque strength not exceeding 40 N cm and maximum speed 15–20 rpm. If the implant was not completely inserted using the contra-angle protocol, insertion was completed using a torque wrench (maximum torque strength 40 N cm) and its screwtool. There was no need to use the screw tap in any case. The implant abutments were then sealed using the cover screws at the top of the implant holder cap. Finally, the gingiva was sutured using Maxon 3–0 absorbable suture thread (Syneture, Tyco Healthcare Group LP, USA). Surgery was performed by experienced implantologists using the techniques commonly used in humans. Treatments and experiments were conducted according to EU guidelines on the welfare of experimental animals. The study protocol was approved by the local ethics committee (application no. 40/ 2005, decision no. 41/2005). At week 12 post-implantation, following sedation with ketamine (10 mg/kg, i.m.) the animals were sacrificed by pentobarbiturate (Thiopental, Biochemie GMbH, Austria) overdose. 2.3. Preparing samples for SEM-BSE
2. Materials and methods 2.1. Oral implants Two types of dental implants available in the market were used: the ‘‘Phoenix’’ implant (Ilerimplant1 S.L.-GMI S.L., Lleida, Spain), a solid screw (outer diameter 4 mm, implanted segment 10 mm) made of c.p. titanium (grade IV) with a
Following animal sacrifice, specimens comprising the dental implants and adjacent host tissue were excised from the mandibular bone as cubes (15 mm 15 mm 15 mm) and immediately transferred to 2.5% glutaraldehyde in 0.1 M phosphate buffer (PB). The samples were chemically fixed in this solution for 24 h at 4 8C. Once fixed, the samples were washed in 0.1 M PB 3 30 min and post-fixed in 1% osmium
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
tetroxide in H2O for 24 h at 4 8C. The specimens were again washed in 0.1 M PB (3 30 min). Next, the samples were dehydrated in a graded ethanol series of 30% for 3 h followed by 50% for 3 h. During dehydration with the next ethanol dilution, the samples were contrasted with saturated 70% ethanol uranyl acetate overnight at 4 8C. This was followed by immersion in 96% (2 3 h) and finally 100% (3 3 h) ethanol. Next, the samples were gradually infiltrated with LRWhite (The London Resin Co. Ltd., Hampshire, UK) embedding medium, first with a 1:1 mixture of LRW in 100% ethanol (3 days at 4 8C) and finally with pure LRW resin (3 3 days at 4 8C). This step was followed by polymerisation in an oxygen-free atmosphere (24 h, 60 8C). After this, the polymerised specimen blocks were visualized by X-ray microradiography to determine the orientation of the implants within the tissue. This helps achieve perfect cross-sectioning through the longitudinal axis of the implant. Next, the blocks were cut using a low speed diamond saw and fine polished using grinding papers (from nr. 300 to 1200) and silicone carbide abrasive sheets with grain diameters of 15, 9 and 3 mm. Final polishing was performed using liquid diamond polishing compound on napped cloth with diamond particle sizes of 1 and 0.25 mm in an oil-based lubricant fluid. The surfaces of the polished cross-sections were post-stained with saturated uranyl acetate (in H2O) for 30 min and with lead citrate (Reynolds, 1963) for 12 min. After washing in distilled water and airdrying, the polished block surfaces were coated with evaporated carbon. 2.4. SEM-BSE microscopy and EDS microanalysis Prepared samples were examined in a scanning electron microscope (DSM940 Zeiss) equipped with a solid-state, four diode BSE detector plus an auxiliary EDS microanalytical system (Link ISIS Oxford). The EDS method involved qualitative and quantitative microanalysis, including element distribution mapping. In the distribution maps provided, relative concentrations are indicated by a colour scale: darkblue represents a concentration of absolute zero and white denotes 100% absolute concentration of the respective purecomponent spectrum (Kotula et al., 2003). The colour concentration scales (0–100%) for each element are provided in the corresponding images. The microscopy and/or microanalytical operating conditions were as follows: 08 tilt angle, 358 take-off angle, 15 kV acceleration potential, 6 or 25 mm working distance and 1–5 nA specimen current. 3. Results The results presented indicate the possible application of the SEM-BSE technique to examine the implant–new bone tissue interface of the specimens prepared using the new procedure. Fig. 1 shows different aspects of the SLA-modified titanium implant–bone interface observed at 12 weeks post-implant. The low magnification view of this interface (Fig. 1a) shows the titanium implant surrounded by newly formed lamellar bone revealing a Haversian, or osteonal bone, pattern. Details of the
1365
Haversian canal, appearing close to the implant surface, are shown in transverse view in Fig. 1b. Note that the wall of this osteon is overlain by a layer of deposited bone tissue and osteoblast cells. According to Skedros et al. (2007), the irregular shape and relatively large diameter of this vascular channel suggests that this is the secondary osteon. The darker contrast of the lamella layer in the SEM-BSE image indicates the lower mean atomic number of the target, probably attributable to the lower content of apatitelike crystals within this structure. In the centre of this canal, a blood vessel containing erythrocytes could be distinguished. Details of the proper integration of the SLA-modified surface implant with the bone are shown in Fig. 1c. A relative abundance of osteocytes was observed in this mineralized mature bone. In some zones of the implant–bone interface, the implant surface was overlain by an extracellular matrix containing bone (osteogenic) cells (Fig. 1d and e). The calcium EDS distribution map (Fig. 1f) clearly demonstrates differences in mineralisation rates (reflected by Ca concentrations) between mineralized bone and immature bone (osteoid), as well as small amounts of calcium in the extracellular matrix. Fig. 2 provides information on the osseointegration process produced in the implants coated with highly porous titanium oxide (TPO). The external micromorphology of this layer (before implant) can be seen in Fig. 2a. Quantitative EDS analysis of this layer revealed a titanium oxide (TiO2) concentration (wt%) of 83.78% (n = 15, S.D. = 1.23). However, this layer was also found to contain a small quantity of phosphorus (the P concentration represented as P2O5 was 16.03%; n = 15, S.D. = 0.37). Our SEM-BSE images (Fig. 2b, c and e) revealed the mineralized bone to be tightly adhered to the TPO layer after 12 weeks of implant. Many of this layer’s conical- and bottle-shaped pores were filled with a matrix composed of calcium, phosphorus and oxygen (EDS map in Fig. 2d). Moreover, the EDS elemental distribution maps (Fig. 2d) distinctly show how the osseous tissue composed of calcium hydroxyapatite adheres to the TPOmodified surface, as well as revealing the appearance of the phosphorous and oxygen elements in the TPO layer (dotted line in Fig. 2d). Fig. 2f shows the occurrence of osteogenic cells close to the implant surface. The SEM-BSE images in Fig. 3 indicate the potential of the proposed strategy to detect and characterize the bone cells present at the implant–neoformed bone interface. Even a low magnification image can provide interesting information on the contents of Haversian canal (Fig. 3a). The detailed view of the centre of the Haversian canal (Fig. 3b) shows osteogenic cells embedded in collagen fibrils. Additionally, the resolution power of this technique can render ultrastructural details of osteocyte-containing lacunae in lamellar bone. Osteocytes are star-shaped cells, and are the most abundant cell type found in bone. The most conspicuous feature of the osteocyte that appears in Fig. 3c is the round nucleus with nucleolus and endoplasmic reticulum. The next image (Fig. 3d) shows mature bone covered with an extracellular matrix of noncalcified bone. Osteoblasts were also frequently observed overlying mature bone (Fig. 3e). In close proximity to these osteoblasts, osteogenic (bone progenitor) cells could also be seen in the
1366
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
Fig. 1. SEM-BSE and EDS cross-section images of the SLA-modified (Phoenix) titanium implant. (a) General view of a titanium (Ti) implant and surrounding bone (B). (b) Detailed view of the area indicated in (a): Ti, implant; B, bone; open arrow, Haversian canal; black arrow, lamella; white arrow, osteoblasts; white arrowhead, erythrocytes. (c) Rough Ti implant surface closely attached to mature mineralized bone (B); white arrows, osteocytes. (d) Osteogenic cells (arrows) and erythrocytes (heads) in the extracellular matrix adjacent to the Ti implant surface. (e) Mineralized bone (B), osteoid (Ot, outlined by a dotted line) and extracellular matrix containing osteogenic cells adjacent to the Ti implant surface. The space (#) between the coating and the Ti implant is an artefact of the tissue preparation process. (f) EDS distribution map showing the calcium (Ca) concentration of the area shown in (e).
extracellular matrix. Odontoblasts (Fig. 3f) were also observed in zones of osseous tissue further from the implant. 4. Discussion The placement of an implant in the alveolar process elicits a sequence of healing events including necrosis and subsequent
resorption of traumatized bone around the titanium fixture concomitant with new bone formation. While the implant initially displays mechanical stability due to contact and friction between the implant surface and the severed bone, the long-term maintenance of implant stability calls for biological attachment between the foreign body and the surrounding tissue. The precision and reliability of a histomorphometric
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
1367
Fig. 2. SEM and EDS cross-section images of the TPO-modified (TiUniteTM) titanium implant. (a) Image obtained by SEM in secondary electron mode showing details of the TPO-modified surface of the TiUniteTM implant. (b) and (c) detailed SEM-BSE micrographs of the interface zone between the Ti implant, TPO-modified implant surface and mineralized bone (B). Note how the bone matrix infills the conical-shaped (open arrow in (b) and bottle-shaped (black arrow in (c) pores of the TPO layer. (d) EDS distribution maps showing Ti, P, Ca and O concentrations, respectively, of the area shown in (c); arrows indicate calcium within the pores of TPO layer. (e) Detailed SEM-BSE view of a bottle-shape pore (boxed area in (c) filled with a Ca–P–O-rich matrix. (f) SEM-BSE micrograph showing a Ti implant and extracellular matrix containing osteogenic cells (arrows) adjacent to the TPO surface.
study of the newly formed bone depends upon the correct identification and ultrastructural characterization of all possible cellular components that play a role in the osseointegration processes. The LM lacks the resolution power required for a detailed structural analysis. The use of electron microscopy techniques could help characterize the morphological changes that occur during osseointegration.
The main electron beam in SEM causes the emission of BSE from the specimen. BSE may be used to detect contrast between areas of different chemical composition. These can be observed especially when the average atomic numbers of the components of the regions vary. Since the intensity of the BSE signal depends on the mean atomic number of the sample (Joy, 1991), the SEMBSE technique not only serves to distinguish inorganic features,
1368
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
Fig. 3. SEM-BSE images of bone cells close to the implant–bone interface. (a) General view of a longitudinal section of a Haversian canal in lamellar bone (B); black arrows, osteocytes; white arrows, osteoblast cells. (b) Detailed view of the area indicated in (a) showing osteogenic cells and collagen fibrils (F). (c) Lacuna occupied by an osteocyte cell in lamellar bone (B); n, nucleus; black arrow, nucleolus; white arrows, endoplasmic reticulum. (d) Mature bone (B) overlain by an extracellular matrix of noncalcified bone (E) with adjacent osteoblast cells; n, nuclei; black arrow, nucleolus. (e) Mature bone (B) overlain by a layer of osteoblast cells (open arrow); black arrows, nucleoli of osteoblast nuclei; white arrows, osteogenic cells. (f) Detailed view of odontoblasts (white arrows, nuclei) and erythrocytes (black arrow) within a blood capillary.
but also offers the interesting possibility of identifying heavy metal-stained ultrastructural cell components and their micromorphological details. The new approach presented in this study was inspired by the successful application of the SEM-BSE technique and heavy metal staining preparation procedure to the study of the microorganism–rock interface. This new research tool was reported first time by Wierzchos and Ascaso (1994). The
present report describes a first attempt at applying this method to characterize the implant–bone tissue interface. SEM-BSE imaging has the advantage of a wide magnification range determining that pieces of bone harbouring implants as small as several centimetres can be scanned. When the site of interest has been identified, it can be magnified and imaged at high-resolution close to that of TEM. Moreover, microanaly-
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
tical determinations in the form of both qualitative and quantitative measurements and element distribution maps can be simultaneously generated for the area of interest. For the stability of a load-bearing implant, the formation of new bone and a mineralized matrix near the implant are key processes. The present study revealed that highly mineralized bone became closely attached to both the SLA- and TPOmodified implant surfaces. Moreover, it was observed (Fig. 2b–e) that the small pores in the TPO-modified implant surface were filled with a Ca–P–O-rich matrix, suggesting the penetration of newly formed bone tissue. This could confer these implants the desired mechanical stability. The presence of small amounts of calcium was also observed in the bone-specific extracellular matrix produced by bone cells concomitant with apatite crystal formation. These findings demonstrating calcification of the bone tissue at the implant–bone interface are in agreement with previous reports, e.g. Simmons et al. (1999). Our ultrastructural data indicate that cells with the typical appearance of welldifferentiated bone cells occur in very close proximity (micrometers) to the implant surface. This observation is consistent with those described by Lavos-Valereto et al. (2001), Simmons et al. (1999) and Meyer et al. (2004) who also noted intimate contact between bone cells and titanium implants on SEM examination in both early and late post-implant stages. 5. Conclusions SEM-BSE visualization of heavy metal-stained bone tissue is very useful for assessing the behaviour of an implant at the bone interface. SEM-BSE images of finely polished crosssections of specimens taken from the contact zone between a titanium implant and surrounding neoformed bone can provide relevant information on the presence and ultrastructural details of bone cells occurring at the implant–bone interface. Moreover, by visualizing the interactions that occur at this interface between cells and their components we will increase our understanding of the osseointegration process. Acknowledgements This research was supported by a formal agreement between the University of Lleida and the firm Ilerimplant S.L., in cooperation with the Kielanowski Institute, the Polish Academy of Sciences and by the Departamento de Universidades, Investigacio´n y Sociedad de la Informacio´n (DURSI) de la Generalidad de Catalun˜a. The authors thank A. Burton for help with preparing the English version of the manuscript and X. Calomarde for help with the sample preparation and technical procedures. The authors are also indebted to J. Esquerda and O. Trabal for their help in interpreting some of the images and J. Qui for X-ray micro-radiography facility. References Bloebaum, R.D., Bachus, K.N., Boyce, T.M., 1990. Backscattered electron imaging: the role in calcified tissue and implant analysis. J. Biomater. Appl. 5, 56–85.
1369
Boyde, A., Jones, S.J., 1996. Scanning electron microscopy of bone: instrument, specimens and tissues. Microsc. Res. Tech. 33, 92–120. Deligianni, D.D., Katsala, N., Ladas, S., Sotiropoulou, D., Amedee, J., Missirlis, Y.F., 2001. Effect of surface roughness of the titanium alloy Ti–6Al– 4V on human bone marrow cell response and on protein adsorption. Biomaterials 22, 1241–1251. Egger, E.L., Gottsauner-Wolf, F., Palmer, J., Aro, H.T., Chao, E.Y.S., 1993. Effects of axial dynamization on bone healing. J. Trauma 34, 185–192. Franch, J., Garcı´a, F., Camo´n, J., Manzanares, M.C., 1998. Backscattered electron imaging of the calcified tissues present in bone healing. Vet. Comp. Orthop. Traumatol. 11, 105–111. Hall, J., Lausmaa, J., 2000. Properties of a new porous oxide surface on titanium implants. Appl. Osseointegr. Res. 1, 5–8. Huang, H.H., Ho, C.T., Lee, T.H., Lee, T.L., Liao, K.K., Chen, F.L., 2004. Effect of surface roughness of ground titanium on initial cell adhesion. Biomol. Eng. 21, 93–97. Jasty, M., Bragdon, C.R., Schtzer, S., Rubash, H., Haire, T., Harris, W.H., 1989. Bone ingrowth into porous coated canine total hip replacements. Quantification by backscattered scanning electron microscopy and image analysis. Scan. Microsc. 3, 1051–1057. Joy, D.C., 1991. An introduction to Monte Carlo simulation. Scan. Microsc. 5, 329–337. Kim, W., Caja, V.L., Larsson, S.A.B., Chao, E.Y.S., 1992. Advanced histomorphometric technique on the analysis of early fracture repair under external fixation. Int. J. Orthop. Trauma Suppl. 3, 5–10. Kotula, P.G., Keenan, M.R., Michael, J.R., 2003. Automated analysis of SEM X-ray spectral images: a powerful new microanalysis tool. Microsc. Microanal. 9, 1–17. Lavos-Valereto, I.C., Wolynec, S., Deboni, M.C., Koing Jr., B., 2001. In vitro and in vivo biocompatibility testing of Ti–6AI–7Nb alloy with and without plasma-sprayed hydroxyapatite coating. J. Biomed. Mater. Res. 58, 727–733. Li, C.Y., Gao, S.Y., Terashita, T., Shimokawa, T., Kawahara, H., Matsuda, S., Kobayashi, N., 2006. In vitro assays for adhesion and migration of osteoblastic cells (Saos-2) on titanium surfaces. Cell Tissue Res. 324, 369–375. Lincks, J., Boyan, B.D., Blanchard, C.R., Lohmann, C.H., Liu, Y., Cochran, D.L., Dean, D.D., Schwartz, Z., 1998. Response of MG63 osteoblast like cells to titanium and titanium alloy is dependent on surface roughness and composition. Biomaterials 19, 2219– 2232. Manzanares Ce´spedes, M.C., Franch Serracanta, J., Carvalho Lobato, P., Montesdeoca Zaeza, J.M., Garcı´a Molina, J.A., Greco Machado, Y., Belmonte Caldero´n, A.M., Delafuente Baigorri, M.P., Serra Renom, I., 2004. The role of calcified tissues in the different stages of implant osteointegration. Dent. Med. Probl. 41, 179–191. Meyer, U., Joos, U., Mythili, J., Stamm, T., Hohoff, A., Fillies, T., Stratmann, U., Wiesmann, H.P., 2004. Ultrastructural characterization of the implant/ bone interface of immediately loaded dental implants. Biomaterials 25, 1959–1967. Ottani, V., Raspanti, M., Martini, D., Tretola, G., Ruggeri Jr., Franchi, M., Giuliani Piccari, G., Ruggeri, A., 2002. Electromagnetic stimulation of bone growth using backscattered electron imaging. Micron 33, 121–125. Reynolds, S., 1963. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell Biol. 17, 200–211. Roschger, P., Fratzl, P., Klaushofer, K., Rodan, G., 1997. Mineralization of cancellous bone after alendronate and sodium fluoride treatment: a quantitative backscattered electron imaging study on minipig ribs. Bone 20, 393–397. Savarino, L., Fini, M., Ciapetti, G., Cenni, E., Granchi, D., Baldini, N., Greco, M., Rizzi, G., Giardino, R., Giunti, A., 2003. Biologic effects of surface roughness and fluorhydroxyapatite coating on osteointegration in external fixation systems: an in vivo experimental study. J. Biomed. Mater. Res. Part A 66, 652–661. Simmons, C.A., Valiquette, N., Pilliar, R.M., 1999. Osseointegration of sintered porous-surfaced and plasma-spray coated implants: an animal model study
1370
J. Wierzchos et al. / Micron 39 (2008) 1363–1370
of early post-implantation healing response and mechanical stability. J. Biomed. Mater. Res. 47, 127–138. Skedros, J.G., Bloebaum, R.D., Bachus, K.N., Boyce, T.M., 1993a. The meaning of greylevels in backscattered electron images of bone. J. Biomed. Mater. Res. 27, 47–56. Skedros, J.G., Bloebaum, R.D., Bachus, K.N., Boyce, T.M., Constantz, B., 1993b. Influence of mineral content and composition on graylevels in backscattered electron images of bone. J. Biomed. Mater. Res. 27, 57–64.
Skedros, J.G., Sorenson, S.M., Jenson, N.H., 2007. Are distributions of secondary osteon variants useful for interpreting load history in mammalian bones? Cell Tissues Organs 185, 285–3007. Sul, Y.T., Johansson, C., Byon, E., Alberktsson, T., 2005. The bone response of oxidized bioactive and non-bioactive titanium implants. Biomaterials 26, 6720–6730. Wierzchos, J., Ascaso, C., 1994. Application of back-scattered electron imaging to the study of the lichen-rock interface. J. Microsc. 175, 54–59.