Addition of human recombinant bone morphogenetic protein-2 to inactive commercial human demineralized freeze-dried bone allograft makes an effective composite bone productive implant material

COMMERCIAL PREPARATIONS OF HUMAN DEMINERALIZED freeze-dried bone allograft (DFDBA) vary in their ability to induce new bone formation. This study test ed the hypothesis that inactive DFDBA can be used as an effective carrier o f recombinant human bone morphogenetic protein-2 (rhBMP-2). Two batches of active DFDBA were used as controls. Two batches of DFDBA, previously shown to be inactive, were treated with vehicle or with 5 or 20 mu g rhBMP-2 and implanted into the calf muscle of male Nu/Nu (nude) mice. Each mouse receiv ed one implant in each hind limb, both of which were of the same formulatio n, resulting in 8 groups of 4 mice per group: active DFDBA batch A, active DFDBA batch B, inactive DFDBA batch A, inactive DFDBA batch B, inactive DFD BA batch A plus 5 mu g rhBMP-2 inactive DFDBA batch A plus 20 mu g rhBMP-2, inactive DFDBA batch B plus 5 mu g rhBMP-2, and inactive DFDBA batch B plu s 20 pg rhBMP-2. After 56 days, the implants were removed and histologicall y examined. A semiquantitative bone induction index was calculated based on the amount of new bone covering each histological section. Histomorphometr y was also used to evaluate the area of new bone formed and the area of res idual implant material. The results showed that active DFDBA induces new bo ne formation, whereas inactive DFDBA does not. Addition of rhBMP-2 to inact ive DFDBA results in new bone formation with a bone induction index compara ble to that of active DFDBA. Histomorphometric analysis, however, revealed that the rhBMP-2-containing implants caused a dose-dependent increase in ne w bone area that exceeded that induced by active DFDBA. At the highest conc entration of rhBMP-2, bone formation was exuberant, rhBMP-2 also caused the resorption of residual implant material to levels comparable to that seen in sites treated with active DFDBA, suggesting that this growth factor may regulate resorptive cells either directly or indirectly. This study shows t hat addition of rhBMP-2 to inactive DFDBA provides reproducible, consistent bone induction, and suggests that inactive commercial preparations may con tain inadequate amounts of BMP to cause bone induction compared to active p reparations.