Porous collagen-glycosaminoglycan (Col/GAG) scaffolds have previously been used clinically as regeneration templates for peripheral nerves and skin^[1]. For defects involving even minimal load-bearing applications however, these scaffolds do not possess the required stiffness. Calcium phosphates (CaPs) are often used as bone-graft substitutes due to their biocompatibility and direct bone-bonding ability. While CaPs have sufficient stiffness for bone-defect applications, unlike Col/GAG they lack elasticity and are very brittle. Combining these two materials produces a composite with enhanced material properties and chemical similarity to natural bone. The addition of CaP nanocrystallites into the Col/GAG matrix produces a 3-dimensional structure that maintains its structural integrity even when wet. In this study, the in vivo performance of mineralised Col/GAG composites was evaluated by implantation into a six-week ovine bone-defect model.

Four different materials were implanted; Col/GAG alone, Col/GAG with octacalcium phosphate, Col/GAG with hydroxyapatite and Col/GAG with brushite. Implants with a diameter of 9mm and length of 9mm, were placed bilaterally into the distal femoral condyle of the hind legs of thirteen sheep. This site was selected due to the large volume of load-bearing cancellous bone. Cancellous autograft was harvested from the tibial tuberosity and placed in the defect sites of two sheep as a positive control.

All animals were sacrificed after 6 weeks and tissue containing the implants was prepared for histological evaluation. Image analysis of Von Kossa stained sections showed that all mineralised Col/GAG implants had significantly more bone in the implant site than unmineralised Col/GAG but were not significantly different between CaPs. Interestingly, new bone formation often followed the structure of the porous material struts which acted as a template. The defect containing the autograft contained the greatest amount of new bone.