A Comparison of In Vivo Bone Tissue Generation Using Calcium Phosphate Bone Substitutes in a Novel 3D Printed Four-Chamber Periosteal Bioreactor

Autologous bone replacement remains the preferred treatment for segmental defects of the mandible; however, it cannot replicate complex facial geometry and causes donor site morbidity. Bone tissue engineering has the potential to overcome these limitations. Various commercially available calcium phosphate-based bone substitutes (Novabone(®), BioOss(®), and Zengro(®)) are commonly used in dentistry for small bone defects around teeth and implants. However, their role in ectopic bone formation, which can later be applied as vascularized graft in a bone defect, is yet to be explored. Here, we compare the above-mentioned bone substitutes with autologous bone with the aim of selecting one for future studies of segmental mandibular repair. Six female sheep, aged 7–8 years, were implanted with 40 mm long four-chambered polyether ether ketone (PEEK) bioreactors prepared using additive manufacturing followed by plasma immersion ion implantation (PIII) to improve hydrophilicity and bioactivity. Each bioreactor was wrapped with vascularized scapular periosteum and the chambers were filled with autologous bone graft, Novabone(®), BioOss(®), and Zengro(®), respectively. The bioreactors were implanted within a subscapular muscle pocket for either 8 weeks (two sheep), 10 weeks (two sheep), or 12 weeks (two sheep), after which they were removed and assessed by microCT and routine histology. Moderate bone formation was observed in autologous bone grafts, while low bone formation was observed in the BioOss(®) and Zengro(®) chambers. No bone formation was observed in the Novabone(®) chambers. Although the BioOss(®) and Zengro(®) chambers contained relatively small amounts of bone, endochondral ossification and retained hydroxyapatite suggest their potential in new bone formation in an ectopic site if a consistent supply of progenitor cells and/or growth factors can be ensured over a longer duration.