In reconstructive surgery, repair of large bone defects is a major reparative problem. About 1,230,000 fractures are treated each year in the USA. Approximately 80 % of these require adjuvant grafting. The most common approaches to this problem utilize autografts and/or allografts from bone banks. Synthetic materials such as hydroxyapatite and tricalcium phosphate ceramics have also been considered. Although autografts are the preferred technique at this time, there are restrictions in its use including donor site morbidity, limited donor bone supply, anatomical and structural problems and elevated level of resorption during healing. Hydroxyapatite (HA) in dense and porous forms is the most widely used material as a bone graft. A constraint in the use of HA is that it is poorly resorbed;​ in addition, the rate of new bone formation on HA is slow.

Because of the limitations associated with biologic and synthetic grafts, we have pursued the idea of using in vitro synthesized bone as graft material. Procedures to synthesize bone tissue in vitro primarily relied on seeding various substrates with cells that have osteogenic capacity in culture. It should be noted that in an in vitro system, osteoprogenitor cells, as well as bone cells themselves can rapidly change their phenotype, hence the substrate needs to promote the expression of the bone cell phenotype. Furthermore, it needs to provide a template for bone deposition while gradually resorbing once bone tissue has been laid down.

The result of this research is a new, porous, surface modified, bioactive glass which acts as a scaffold stimulating bone formation in vitro. The porosity of the glass was 36.4%;​ the pore size ranged from 10 to 160 μm, and there was no incipient devitrification. Because BG undergoes corrosion in an aqueous environment, the effect of alkali ions released, in vitro, from the surface of porous bioactive glass (BG) on the cellular activity of neonatal rat calvaria osteoblasts was studied. At low tissue culture medium (TCM) volume to BG surface area (SA) (Vol/SA), the glass corrosion products shift the pH of the TCM to a value that adversely affects the cellular activity. Cells cultured on BG at low Vol/SA conditions produced non mineralized extracellular matrix (ECM). On the other hand, mineralized ECM was produced by cells seeded on BG and maintained at high Vol/SA.

Since the overall goal of our study is to be able to implant bone-like tissue, synthesized in vitro, into defect sites, it was critical that the osteoblastic response be maximized so that there is an abundant and expeditious formation of bone in vitro. We found it was necessary to condition the disks to be covered with a dual layer of calcium phosphate and serum protein, prior to seeding the glass with cells. We evaluated the importance of the presence of a surface crystalline versus amorphous calcium phosphate layer on the activity of osteoblasts seeded on bioactive glass. Furthermore, the effect on bone cell function of serum protein adsorbed on modified bioactive glass and hydroxyapatite ceramic was also studied. Optimum cellular activity was observed on BG conditioned to be covered with crystalline calcium phosphate-serum protein layer.

Immersing the glass in simulated body fluid for 180 h leads to the formation of a crystalline calcium phosphate layer at the glass surface. This layer is a carbonated hydroxyapatite. The layer provides sites for attachment of growth factors and adhesion factors such as fibronectin present in serum of the TCM. Concentration of these factors at the glass surface serves to enhance the expression of the osteoblast phenotype and the production of significant amount of bone-specific proteins. We found that the modified glass templates were rapidly invaded by cells which maintained the osteoblast phenotype;​ thus, they exhibited high alkaline phosphatase activity, they synthesized type I and m collagen, osteocalcin, osteopontin, and bone sialoprotein. SEM-EDAX showed that the cells elaborated substantial amounts of extracellular matrix and a bone-like tissue was present throughout the entire template thickness. FTER. analysis of material formed in the glass indicated that the mineral phase was a biological hydroxyapatite. In contrast a much smaller number of cells attached to the HA ceramic surface;​ although these cells synthesized similar profile of proteins, they produced a limited mineralized matrix. On the bioactive glass surface, the high selectivity of the calcium phosphate layer for fibronectin may be responsible for the bone cellular activity that is observed in vitro as well as when this material is implanted in vivo.