In vitro model for frontal sinus obliteration with bioactive glass S53P4

An in vitro model was used to investigate the behavior of a massive frontal sinus obliteration with bioactive glass S53P4 (BG) for clinical purposes. Two sizes of granules (0.63-0.8 mm or 0.8-1.0 mm) in 16 separate BG amounts, weight 25 g, were tested both in simulated body fluid (SBF) and a buffer containing trishydroxymethyl aminomethane citric acid (TRIS-c.a) in standard conditions. The dissolution of silicon (Si) and phosphate (P) was detected with direct current plasma atom emission spectroscopy (DCP-AES) monthly up to 6 months. The BG masses were scanned by computer tomography (CT) and the scans analyzed by Region of Interest (ROI) technique. Calcium phosphate (CaP)- and silica (Si)-gel-layers were studied by scanning electron microscopy (SEM) at 1, 3, and 6 months. Cumulative loss of Si and P was stronger in TRIS -c.a than in SBF (p < 0.0001), and it was higher with smaller than with larger granules in both solutions (p < 0.0001). This was shown correspondingly by the decrease in Hounsfield units (HU) by ROI analysis (p < 0.0001). In SBF-soaked BG masses, the CaP-layer occurred on the uppermost granules, and in TRIS-c.a at 3-6 months, on the granules in the center and lower parts. The decrease of HU seems to reveal indirectly the resorption of BG.     

Clinical follow-up method for frontal sinus obliteration with bioactive glass S53P4

A clinical follow-up method was developed to investigate the behavior of a massive amount of bioactive glass S53P4 (BG) clinically used in frontal sinus obliteration. Two sizes of granules (0.63-0.8 mm or 0.8-1.0 mm) in 16 separate BG amounts, weight 25 g, were tested both in simulated body fluid (SBF) and in a buffer containing tris-hydroxymethyl aminomethane citric acid (TRIS-c.a) in standard conditions. The dissolution of silicon (Si) and phosphate (P) was detected with direct current plasma atom emission spectroscopy (DCP-AES) monthly up to 6 months. The BG masses were scanned both wet in the solutions and dried by computer tomography (CT), and the scans were analyzed by Region of Interest (ROI) technique. Calcium phosphate (CaP)- and silica (Si)-gel-layers were studied by scanning electron microscopy (SEM) at 1, 3, and 6 months. Cumulative loss of Si and P was stronger in TRIS-c.a than in SBF (p < 0.0001), and it was higher with smaller than with larger granules in both solutions (p < 0.0001). This was shown correspondingly by the decrease of Hounsfield units (HUs) in ROI analysis (p < 0.0001). The level of HUs was lower with dried than with wet BG (p < 0.0001). The results were compared for clinical ROI analysis of patients with obliterated frontal sinuses up to 48 months and they were parallel. The follow-up method seems to indirectly reveal the behavior of BG and the healing process in the obliterated cavity.     

Interactions between the bioactive glass S53P4 and the atrophic rhinitis-associated microorganism klebsiella ozaenae

In an aqueous environment, ions are released from a bioactive glass (BAG) and the pH rises in its vicinity. This may influence both growth and colonization of microorganisms. We studied the effects of the BAG S53P4 on the atrophic rhinitis-associated microorganism Klebsiella ozaenae. The glass was used in the form of granules or discs. Growth inhibition was studied using an agar plate test. Adhesion was studied by incubating bacterial suspension with the glass. The effect of the presence of the bacteria on the formation of the Si-rich layer on the bioactive glass was also analyzed. Furthermore, a follow up study of 19-74 months with ozena patients surgically treated with the BAG S53P4 was performed. The bioactive glass showed no clear growth inhibition of K. ozaenae in the agar plate test. K. ozaenae showed low adherence to the BAG S53P4. No growth of the microbe was seen on the glass during the 8 h incubations and the Si-rich layer was formed normally. The clinical follow-up study showed no infections of the implants and the symptoms of the patients were markedly reduced. Thus, the BAG S53P4 did not favor adhesion and colonization of K. ozaenae, in vitro, which is supported by the in vivo findings showing no BAG-associated infections or reinfections.     

Phase composition and in vitro bioactivity of porous implants made of bioactive glass S53P4

This work studied the influence of sintering temperature on the phase composition, compression strength and in vitro properties of implants made of bioactive glass S53P4. The implants were sintered within the temperature range 600-1000°C. Over the whole temperature range studied, consolidation took place mainly via viscous flow sintering, even though there was partial surface crystallization. The mechanical strength of the implants was low but increased with the sintering temperature, from 0.7 MPa at 635°C to 10 MPa at 1000°C. Changes in the composition of simulated body fluid (SBF), the immersion solution, were evaluated by pH measurements and ion analysis using inductively coupled plasma optical emission spectrometry. The development of a calcium phosphate layer on the implant surfaces was verified using scanning electron microscopy-electron-dispersive X-ray analysis. When immersed in SBF, a calcium phosphate layer formed on all the samples, but the structure of this layer was affected by the surface crystalline phases. Hydroxyapatite formed more readily on amorphous and partially crystalline implants containing both primary Na(2)O·CaO·2SiO(2) and secondary Na(2)Ca(4)(PO(4))(2)SiO(4) crystals than on implants containing only primary crystals.     

Compound changes and tooth mineralization effects of glass ionomer cements containing bioactive glass (S53P4), an in vivo study

In this study, modifications of glass ionomer cements (GICs) were made by adding bioactive glass (BAG) to GIC to obtain bioactive restorative materials. This study used SEM, EDS and visual analysis to examine the bioactivity and the ability of the study materials to mineralize dentin. Conventional cure and resin-modified light-curing GIC were used. The materials consisted of powder and liquid. Three experimental materials were made by mixing 10-30 wt% of BAG powder with GIC powders. Commercially available GIC without BAG were used as controls. Class III restorations were made in altogether 62 intact beagle dog teeth, and the operation was performed under general anesthesia. The restorations were followed clinically for 1, 3 or 6 weeks. Resin-modified GIC containing BAG showed uniform CaP surface formation on the restorations. Mineral depositions in the close vicinity of the restoration-dentin interface and in deeper parts of dentin tubules were also noticed in resin-modified GIC containing BAG particles. It can be concluded that resin-modified GIC containing BAG have good potential in clinical applications where enhanced mineralization is expected.     

In vivo model for frontal sinus and calvarial bone defect obliteration with bioactive glass S53P4 and hydroxyapatite

An in vivo model was developed to investigate the usability of a frontal sinus and a calvarial bone defect obliteration with bioactive glass S53P4 (BG) and hydroxyapatite (HA) granules. Roofs of 21 Elco rabbit frontal sinuses were drilled open from 4 separate holes using a standard method, and the sinuses, located in pairs, in frontal bone were filled with BG on one side and with HA on the other side. Two parallel posterior defects were covered with a pedicled periosteum flap, and 2 anterior defects with a free flap. The stability of materials, new bone, and connective tissue formation were observed with histomorphometry, scanning electron microscopy (SEM), energy dispersive X-ray analysis (EDXA), and X-ray pictures at 1, 3, 6, and 12 months postoperatively. The results showed more rapid resorption of filling material (p = 0.019) and new bone formation (p = 0.0001) in the defects filled with BG than in the corresponding HA-filled defects studied by histomorphometry throughout the study. New bone formation and resorption of materials were faster in defects covered by a pedicled flap than by a free periosteum flap. The results were supported by SEM histomorphometric and radiologic analysis. Both bioactive materials studied were well tolerated in frontal sinuses and in calvarial bone defects. The experimental model showed the influence of early periosteum vascularization on accurate frontal sinus filling and the healing process in rabbit frontal sinuses.     

In vitro Ca-P precipitation on biodegradable thermoplastic composite of poly(epsilon-caprolactone-co-DL-lactide) and bioactive glass (S53P4)

Bioactive properties of composites containing poly(epsilon-caprolactone-co-DL-lactide) with molar ratio 96/4 and bioactive glass (BAG), S53P4, were tested in vitro. The glass content in the tested materials was 40, 60 or 70 wt%, and two granule size ranges (<45 and 90-315 microm) were used. The composites were analysed for their apatite-forming ability. This was determined as a function of time by the dissolution pattern of Si and Ca ions and structural changes on the specimen surfaces. Composite specimens were immersed in simulated body fluid at 37 degrees C for up to 6 months. The changes in Si and Ca concentrations of the immersion medium were determined with UV-Vis and atomic absorption spectrophotometry. The calcium phosphate precipitation and apatite formation were evaluated by scanning electron microscopy (SEM) and infra-red spectroscopy (IR) using the attenuated total reflectance (ATR) system. The SEM and SEM-EDX analysis of the depositions formed on the composite surfaces was in line with the changes in ion concentrations. The clearest results with IR were seen in the material containing 60 wt% small glass particles. The results indicate that composites containing over 40 wt% BAG granules are bioactive, and that a higher BAG surface area/volume ratio favors the apatite formation in vitro.     

Matrix collagen of devitalized bone is resistant to osteoclastic bone resorption

The resorption of devitalized bone by isolated osteoclasts in culture was studied by scanning electron microscopy. Osteoclasts attached to the bone and resorbed mineral but left most of matrix collagen undigested in the resorption pits. Postculture-treatment of bone substrate with collagenase completely removed the matrix collagen in the resorption pits, whereas trypsin did not. The results suggest that native matrix collagen is resistant to osteoclastic attack in this in vitro model using devitalized bone.     

Sintering behaviour of 45S5 bioactive glass

In this study, we report on the effect of Bioglass structural transformations on its sintering behaviour. While heating up to 1000 degrees C, five successive transformations occur: glass transition, glass-in-glass phase separation, two crystallization processes and a second glass transition. The sintering of the material exhibits two main shrinkage stages associated with the two glass transitions at 550 and 850 degrees C. At 580 degrees C, the glass-in-glass phase separation induces a decrease in the sintering rate immediately followed by the major crystalline phase crystallization (Na(2)CaSi(2)O(6)) between 600 and 700 degrees C, from the surface to the bulk of the particles. A complete inhibition of sintering takes place followed by a minor shrinkage effect due to crystallization. A plateau is then observed until the second glass transition temperature is reached. A modification of Frenkel's model allows the determination of the glass-in-glass phase separation kinetics and the identification of the structural transformations effects on sintering behaviour.     

Creation of microrough surface on sintered bioactive glass microspheres

Bioactive glasses are surface-active, generally silica-based, synthetic materials that form a firm chemical bond to bone. The aim of this study was to further enhance the bioactivity of glasses by creating a microroughness on their surface. Microroughness increases potential surface area for cell attachment and biomaterial-cell interactions. Three bioactive glasses of different composition were studied. Each material was flame-sprayed into microspheres, and a selected fraction of the spheres (250-300 microm) was sintered to form porous bioactive glass specimens. To create microrough surfaces, different acid etching techniques were tested. Atomic force microscopy (AFM) and back-scattered electron imaging of scanning electron microscopy (BEI-SEM) were used to characterize surface roughness. The degree of roughness was measured by AFM. A novel chemical-etching method, developed through intensive screening of different options, was found consistently to create the desired microroughness, with an average roughness value (R(a)) of 0.35-0.52 microm and a root mean-square roughness value (R(rms)) of 0.42-0.64 microm. Microroughening of the glass surface was obtained even in the internal parts of the porous glass matrices. Measured by BEI-SEM, the etching of a bioactive glass surface did not interfere with the formation of the characteristic surface reactions of bioactive glasses. This was confirmed by immersing the etched and control glass bodies in a simulated body fluid and tris(hydroxymethyl) aminomethane/HCl. The etching process did not significantly affect the mechanical strength of the sintered bioactive glass structures. Based on these experiments, it seems possible to create a reproducible microroughness of appropriate size on the surface of porous bioactive glass. The biologic benefits of such a surface treatment need to be validated with in vivo experiments.     

Use of glass slides coated with apatite-collagen complexes for measurement of osteoclastic resorption activity

This study was designed to evaluate the use of apatite-collagen complexes (ACC) coated onto glass slides for measurement of osteoclastic resorption activity. ACC-coated glass slides were prepared by immersion in beta-glycerophosphate solution for 7-14 days after glass slides coated with type I collagen had been treated with alkaline phosphatase and phosvitin. Osteoclast-containing cell suspensions were prepared from the long bones of 1-day-old rabbits and were seeded in medium 199 (containing 10% FBS) onto ACC-coated glass slides. After allowing the cells to attach for 1.5 h, the glass slides were incubated for periods of up to 96 h. The cells were observed by scanning electron microscopy and cytochemically for tartarate resistant acid phosphatase (TRAP) activity. Some slides were treated with FITC-phalloidin and anti-type I collagen antibody. TRAP-positive multinucleated cells were located in transparent spaces on the glass slides. These spaces did not stain immunohistochemically with anti-type I collagen antibody. Podosome formation was observed in the multinucleated cells facing the edge of the transparent spaces. The scanning electron microscopy demonstrated well-spread large cells located on the flattened surface on apatite particles covering the glass surface. Our results suggest that osteoclasts could resorb the apatite particles and coated collagen on the glass slide. The resorption lacunae appeared as transparent spaces, and the cytoskeleton of resorbing osteoclasts was observed in these spaces. ACC-coated glass slides could be useful for investigating the function and metabolic activities of osteoclasts.     

Calcium phosphate formation at the surface of bioactive glass in vitro

The calcium phosphate formation at the surface of bioactive glass was studied in vitro. Glass rods and grains were immersed in different aqueous solutions and studied by means of scanning electron microscopy and energy dispersive x-ray analysis. Surface morphological changes and weight loss of corroded grains were monitored. In-depth compositional profiles were determined for rods immersed in the different solutions. The solutions used were tris-buffer (tris-hydroxymethylaminomethane + HCl), tris-buffer prepared using citric acid (tris-hydroxymethylaminomethane + C6H8O7.H2O), and a simulated body fluid, SBF, containing inorganic ions close in concentration to those in human blood plasma. It was found that the calcium phosphate formation at the surface of bioactive glass in vitro proceeds in two stages. When immersing the glass in tris or in SBF a Ca,P-rich surface layer forms. This accumulation takes place within the silica structure. Later, apatite crystals forming spherulites appear on the surface. The Ca/P-ratio of initially formed calcium phosphate was found to be about unity. It is proposed that this is due to bonding of phosphate to a silica gel. The surface is stabilized, i.e., leaching is retarded, by the rapid Ca,P-accumulation within the silica structure before apatite crystals are observed on the surface. It is proposed that the initially formed calcium phosphate is initiated within the silica gel. The crystallizing surface provides nucleation sites for extensive apatite formation on the glass surface. In the presence of citrate no Ca,P-accumulation occur at the glass surface, but soluble Ca-citrate complexes form. By comparing the weight loss during corrosion in tris with that in the calcium and phosphate containing SBF, it is possible to establish whether the glass can induce apatite formation at its surface or not.     

Characterization of microrough bioactive glass surface: surface reactions and osteoblast responses in vitro

The current study characterized the in vitro surface reactions of microroughened bioactive glasses and compared osteoblast cell responses between smooth and microrough surfaces. Three different bioactive glass compositions were used and surface microroughening was obtained using a novel chemical etching method. Porous bioactive glass specimens made of sintered microspheres were immersed in simulated body fluid (SBF) or Tris solutions for 1, 6, 24, 48, or 72 h, and the formation of reaction layers was studied by means of a scanning electron microscope/energy dispersive X-ray analysis (SEM/EDXA). Cell culture studies were performed on bioactive glass disks to examine the influence of surface microroughness on the attachment and proliferation of human osteoblast-like cells (MG-63). Cell attachment was evaluated by means of microscopic counting of in situ stained cells. Cell proliferation was analyzed with a nonradioactive cell proliferation assay combined with in situ staining and laser confocal microscopy. The microroughening of the bioactive glass surface increased the rate of the silica gel layer formation during the first hours of the immersion. The formation of calcium phosphate layer was equal between control and microroughened glass surfaces. In cell cultures on bioactive glass, the microrough surface enhanced the attachment of osteoblast-like cells but did not have an effect on the proliferation rate or morphology of the cells as compared with smooth glass surface. In conclusion, microroughening significantly accelerated the early formation of surface reactions on three bioactive glasses and had a positive effect on initial cell attachment.     

45S5 bioactive glass surface charge variations and the formation of a surface calcium phosphate layer in a solution containing fibronectin

This study investigated the effect of fibronectin adsorption on surface charge variations and calcium phosphate (Ca-P) layer formation kinetics on the surface of 45S5 bioactive glass (BG). We hypothesize that the adsorption of fibronectin on BG changes the surface charge and alters the kinetics of Ca-P layer formation on the glass surface. The charge at a material's surface modulates surface chemistry, protein adsorption, and interactions with bone cells. The zeta potential of BG in a solution containing human plasma fibronectin (TE-FN) was measured as a function of time by particle electrophoresis, and Ca-P layer formation was characterized using SEM, EDXA, and FTIR. Si, Ca, and P solution concentrations also were determined. It was found that the adsorption of fibronectin reduced the initial electronegativity of the BG surface and delayed the formation of both the amorphous and the crystalline Ca-P layers. The delayed formation of these surface layers may be attributed to the competitive binding of Ca2+ ions by the fibronectin molecule. In addition, the formation of an amorphous Ca-P layer correlated with the reversal from a negatively to a positively charged surface, independent of the presence of fibronectin. The addition of a single protein (in this case fibronectin) can significantly alter material surface parameters, such as charge, and subsequently affect the formation of a surface Ca-P layer. Furthermore, the formation of an amorphous Ca-P layer is an important event in the reactions leading to bioactive behavior, and proteins such as FN are actively involved in the transformation of the surface into a Ca-P layer.     

Bioactive glass S53P4 in repair of septal perforations and its interactions with the respiratory infection-associated microorganisms Haemophilus influenzae and Streptococcus pneumoniae

Interpositional grafts between mucoperiosteal flaps are commonly used in the repair of septal perforations. We studied the use of bioactive glass (BAG) S53P4 as an interpositional graft in 11 patients suffering from septal perforations. In aqueous environments, ions are released from the BAG and the pH rises in its vicinity, both of which may influence the growth and adhesion of microorganisms. Thus, we also studied the effects of the BAG S53P4 as granules or discs on the respiratory infection-associated microorganisms Haemophilus influenzae and Streptococcus pneumoniae. Growth inhibition was studied using an agar plate test and adhesion was analyzed both with and without serum precoating of the BAG S53P4. The perforations were successfully closed in 10 of 11 patients. One patient had a near total septum perforation, which could not be closed. No BAG-associated infections were seen during the follow-up. The BAG S53P4 did not show any clear growth inhibition of the microorganisms, which showed low adhesion to the material. Serum precoating increased the adsorption. Thus, uncoated BAG S53P4 seems to be a good graft in the repair of septal perforations.     

Effect of varying physical properties of porous, surface modified bioactive glass 45S5 on osteoblast proliferation and maturation

The objective of this study was to determine the effect of porous bioactive glass (45S5) substrate characteristics on the expression and maintenance of the osteoblastic phenotype. We cultured ROS 17/2. 8 cells on substrates with different pore size and porosity for periods up to 14 days and analyzed the characteristics of the cells and extracellular matrix. Results of the study show that the glass substrates supported the proliferation and growth of osteoblast-like cells. Although the morphologies of the cells differed on the various substrates, their shape and the extent of membrane ruffling suggested that they maintained high levels of metabolic activity. Cells on all substrates expressed high levels of alkaline phosphatase activity and produced extracellular matrices that mineralized to form nonstoichiometric, carbonated, calcium-deficient apatites. An important finding was that at a given porosity of 44%, the pore size neither directed nor modulated the in vitro expression of the osteoblastic phenotype. In contrast, porosity did affect cellular function. We noted that at an average pore size of 92 microm, as the porosity increased from 35 to 59%, osteoblast activity was reduced. As designed in this experiment, an increase in the porosity led to a corresponding increase in total surface area of the specimens. With increasing porosity and surface area, glass reactions in the media may persist for longer durations at higher intensities, thereby affecting local media composition. As such, we suggest that extensive conditioning treatments before cell seeding can reduce this effect. Our results also revealed that the expression of the osteoblastic phenotype is enhanced by the ongoing glass dissolution. The reaction pathway at the origin of this effect still needs to be elucidated. Taken together, the findings support the overall hypothesis that in vitro cell activity can be controlled by a careful selection of substrate properties.     

Biologic significance of surface microroughing in bone incorporation of porous bioactive glass implants

A novel chemical etching method was recently developed to create a controlled microrough surface on porous bioactive glass implants. Our earlier in vitro studies showed enhanced attachment of osteoblast-like MG63 cells on a microrough bioactive glass surface. The purpose of our current study was to confirm the in vivo significance of surface microroughening for bone bonding of bioactive glass. Porous bioactive glass cones made of sintered microspheres were surgically implanted in the anterior cortex of rabbit femurs. Peripheral quantitative computed tomography (pQCT), biomechanical push-out testing, histomorphometry, and electron microscopy (BEI-SEM) were used to analyze bone ingrowth and osseointegration at 7, 10, 14, 28, 56, and 84 days after implantation. The results showed that microroughening of the bioactive glass surface significantly enhanced the bone-bonding response of the biomaterial. The positive response was seen in one of the three bioactive glass compositions studied. The affinity index of new bone on the glass surface was significantly (p = 0.02) increased with a trend (p = 0.10) toward improved mechanical incorporation. New bone formation was dependent on the glass composition, and it was found to occur not only through the mechanism of bone ingrowth but also based on in situ osteogenesis within implant interstices. Based on these results, the procedure of microroughening could enhance the osteopromotive properties of certain bioactive glass compositions.     

Influence of hydroxyl content on selected properties of 45S5 bioactive glass

Numerous material properties may be influenced by the concentration of chemically dissolved hydroxyl species within a glass. A tube furnace connected to a steam generator was used to create hydroxyl-saturated 45S5 glass under 1 atm of water at 1100 degrees C. Selected properties of as-melted and hydroxyl-saturated samples were compared to assess the sensitivity of 45S5 to excess hydroxylation. The glass transition temperature and the peak crystallization temperature of the treated 45S5 glass were reduced in comparison to the as-melted 45S5 glass. In addition, the treated glass exhibited a broad endothermic signal that may be indicative of enhanced viscous flow. A simple dissolution experiment indicated that the treated 45S5 glass was also less durable than the as-melted 45S5 glass.     

Review of bioactive glass: From Hench to hybrids

Bioactive glasses are reported to be able to stimulate more bone regeneration than other bioactive ceramics but they lag behind other bioactive ceramics in terms of commercial success. Bioactive glass has not yet reached its potential but research activity is growing. This paper reviews the current state of the art, starting with current products and moving onto recent developments. Larry Hench’s 45S5 Bioglass® was the first artificial material that was found to form a chemical bond with bone, launching the field of bioactive ceramics. In vivo studies have shown that bioactive glasses bond with bone more rapidly than other bioceramics, and in vitro studies indicate that their osteogenic properties are due to their dissolution products stimulating osteoprogenitor cells at the genetic level. However, calcium phosphates such as tricalcium phosphate and synthetic hydroxyapatite are more widely used in the clinic. Some of the reasons are commercial, but others are due to the scientific limitations of the original Bioglass 45S5. An example is that it is difficult to produce porous bioactive glass templates (scaffolds) for bone regeneration from Bioglass 45S5 because it crystallizes during sintering. Recently, this has been overcome by understanding how the glass composition can be tailored to prevent crystallization. The sintering problems can also be avoided by synthesizing sol–gel glass, where the silica network is assembled at room temperature. Process developments in foaming, solid freeform fabrication and nanofibre spinning have now allowed the production of porous bioactive glass scaffolds from both melt- and sol–gel-derived glasses. An ideal scaffold for bone regeneration would share load with bone. Bioceramics cannot do this when the bone defect is subjected to cyclic loads, as they are brittle. To overcome this, bioactive glass polymer hybrids are being synthesized that have the potential to be tough, with congruent degradation of the bioactive inorganic and the polymer components. Key to this is creating nanoscale interpenetrating networks, the organic and inorganic components of which have covalent coupling between them, which involves careful control of the chemistry of the sol–gel process. Bioactive nanoparticles can also now be synthesized and their fate tracked as they are internalized in cells. This paper reviews the main developments in the field of bioactive glass and its variants, covering the importance of control of hierarchical structure, synthesis, processing and cellular response in the quest for new regenerative synthetic bone grafts. The paper takes the reader from Hench’s Bioglass 45S5 to new hybrid materials that have tailorable mechanical properties and degradation rates.     

The in vitro response of osteoblasts to bioactive glass

Osteoblasts from neonate rat calvaria migrated in culture from the endocranial surface of parietal bones onto fragments of bone-bonding 45S5 glass or non-bone-bonding quartz glass. These organ culture units were maintained for up to 4 wk. No significant production of extracellular matrix (ECM) was seen on the quartz glass samples. However, osteoblasts colonized the 45S5 samples in multilayers and produced abundant ECM as seen by light (LM), scanning electron (SEM) and transmission electron (TEM) microscopy. The interface developed on 45S5 glass was designated as either Type I (non-collagen-bonding) or Type II (showing direct interdigitation of collagen with the calcium phosphate-rich glass surface). It was concluded that, since this in vitro method is capable of reproducing some aspects of the known in vivo behaviour of 45S5, such techniques may be developed as a means of batch-testing bioactive biomaterials and investigating bone cell/biomaterial interactions.