Magnesium (Mg) is an emerging degradable metal for orthopaedics and its degradation product, Mg²⁺, has been reported to positively affect osteogenesis. Porous hydroxyapatite (HA) has been extensively studied for bone regeneration, but its slow degradation is an important factor limiting its uses. Therefore, we studied the combination of Mg and HA to integrate the advantages of the two materials while circumventing their disadvantages, with the aim of determining the optimum Mg/HA ratio. Surface-treated Mg particles (Φ ∼ 1.3 mm) and HA spheres (Φ ∼ 1.4 mm) were mixed at 0, 10, 30, and 50% (v/v). The mixtures were filled in titanium cages to form complexes. In vitro biocompatibility was studied by culturing MC3T3-E1 osteoblasts in extracts prepared from the complexes. In vivo biocompatibility and osteoinduction were evaluated by implantation in canine dorsal muscles. All complexes were not cytotoxic to the osteoblasts, regardless of the Mg proportion. Eight weeks after implantation in muscles, complexes containing 50% Mg showed inflammatory cell infiltration, whereas all other groups showed no evident inflammatory response. Additionally, complexes containing 30% Mg had significantly higher new bone area percentage (BV/TV), trabecular number (Tb.N), and bone growth rate than those containing 0, 10%, or 50% Mg. In comparison, complexes containing 50% Mg had significantly lower BV/TV and Tb.N than all other groups, and also had a significantly lower bone growth rate than those containing 10% or 30% Mg. Sixteen weeks after implantation, the trend in bone formation was similar to that observed at week 8. This shows that metallic Mg affected the osteoinductivity of HA–Mg complexes in a dose-dependent manner. In the present model, 30% Mg induced optimum osteoinduction. More studies are needed to explore the optimum Mg fractions for other implantation sites or forms of use.

Porous hydroxyapatite (HA) is widely used in porous forms to assist bone defect healing. However, further improvements in biological functions are desired for meeting complex clinical situations such as impaired bone regeneration in poor bone stock. The extracellular matrix (ECM) of human tissues is characterized by nanofibrous structures and a variety of signal molecules. Emulating these characteristics are expected to create a favorable microenvironment for cells and simultaneously allow release of osteogenic molecules. In this study, short polylactide fibers containing BMP-2 were prepared by electrospinning and coated on porous HA scaffolds. The coating did not affect porosity or pore interconnectivity of the scaffold but improved its compressive strength markedly. This fiber coating produced burst BMP-2 release in 1 day followed by a linear release for 24 days. The coating had a significantly lower rat calvarial osteoblasts (RCOBs) adhesion (vs. uncoated scaffold) but allowed normal proliferation subsequently. Bone marrow stem cells (MSCs) on the coated scaffolds expressed a significantly increased alkaline phosphatase activity relative to the uncoated ones. After implantation in canine dorsal muscles, the coated scaffolds formed significantly more new bone at Weeks 4 and 12, and more blood vessels at Week 12. This method offers a new option for drug delivery systems.