Healing bone using recombinant human bone morphogenetic protein 2 and copolymer.

Middiaphyseal 2.5-cm segmental defects in the right femurs of 12 sheep were stabilized with stainless steel plates and implanted with (1) 2 mg recombinant human bone morphogenetic protein 2 and poly[D,L-(lactide-co-glycolide)] bioerodible polymer with autologous blood (n = 7), (2) 4 mg recombinant human bone morphogenetic protein 2 and poly[D,L-(lactide-co-glycolide)] and blood (n = 3), or (3) poly[D,L-(lactide-co-glycolide)] and blood only (n = 2). Bone healing was evaluated for 1 year using clinical, radiographic, gross pathologic, and histologic techniques. Union occurred in three sheep in Group 1, two in Group 2, and none in Group 3. In the animals that healed, new bone first was visible radiographically between Weeks 2 and 6 after implantation; new bone mineral content equaled that of the intact femur not surgically treated by Week 16; recanalization of the medullary cavity approached completion at Week 52; and at necropsy the surgical treated femurs were rigidly healed, the poly[D,L-(lactide-co-glycolide)] was resorbed completely, and woven and lamellar bone bridged the defect site. In two Group 1 sheep euthanized at Weeks 2 and 6, polymer particles were permeated by occasional multinucleated giant cells. Some plasma cells, lymphocytes, and neutrophils were present locally. The poly[D,L-(lactide-co-glycolide)] tended to fragment during surgical implantation. Despite these observations, the recombinant human bone morphogenetic protein 2/poly[D,L-(lactide-co-glycolide)] implant was able to heal large segmental bone defects in this demanding model.