Comparison of two carbonated apatite ceramics in vivo

Carbonated apatite ceramics, with a composition similar to that of bone mineral, are potentially interesting synthetic bone graft substitutes. In the present study, two porous carbonated apatite ceramics were developed, characterized and tested for their bone repair capacity and osteoinductive potential in a goat model. Although the two ceramics were prepared from a similar starting powder, their physico-chemical and structural characteristics differed as a consequence of different preparation methods. Both ceramics had an open and interconnected porous structure with a porosity of about 80%. The morphology of the surface of CA-A and CA-B at the submicron level differed significantly: CA-A consisted of irregular grains with a size of 5–20 μm, with 1–10 μm large micropores among the grains, whereas CA-B surface consisted of much smaller and regular shaped grains (0.05–0.5 μm), with most micropores smaller than 1 μm. The specific surface area of CA-B was about 10 times larger than that of CA-A due to its significantly smaller grain size. CA-A and CA-B ceramics contained 3 and 5 wt.% of B-type carbonated apatite, respectively. Although neither ceramic succeeded in completely bridging the 17 mm iliac wing defect with new bone after 12 weeks of implantation, CA-A showed significantly more bone formation in the pores of the implant than CA-B. The total area percentage of new bone in the total defect area was 12.7 ± 1.81 and 5.51 ± 1.37 (mean ± SEM) for CA-A and CA-B, respectively. Intramuscular implantation of the ceramics led to ectopic bone formation by CA-A in all three implanted specimens, in contrast to CA-B, where no new bone was observed in any of the 11 animals. CA-A showed a more pronounced degradation than CA-B both in vitro and in vivo at both implantation sites, which was unexpected based on the physico-chemical and structural properties of the two ceramics. Both physico-chemical and structural properties of the ceramics could, dependently or independently, have affected their in vivo behaviour, emphasizing the importance to control individual parameters for successful bone repair.