Early reconstruction of large osseous defects in children is often delayed due to limited availability of autogenous bone graft donor sites. With the advent of growth factors, osteoinductive proteins, and delivery matrices, it is possible to fabricate new bone at extraskeletal sites. Due to their own blood supply, adequate bony volume, and decreased resorption, vascularized bone flaps have demonstrated greater success in restoring large bony defects compared with nonvascularized bone grafts. The purpose of this study is to prefabricate a vascularized bone flap in the immature-age rabbit using the auricularis anterior muscle as a muscle pedicle. Sixteen female New Zealand White rabbits, 2.0 to 2.5 kg, were divided into two groups. Group 1 contained 8 animals that had T-shaped, 10 x 6 x 4-mm hydroxyapatite (HA) implants combined with 100-microgram bovine-derived bone morphogenetic protein (BMP) placed supraperiosteally and fixed deep to the auricularis anterior muscle. Implants with HA alone were placed in the same animal and secured to the contralateral auricularis anterior muscle. Group 2 contained 8 animals that had HA/BMP placed subperiosteally and fixed deep to the auricularis anterior muscle, while implants with HA alone were secured in the same animal to the contralateral auricularis anterior muscle. In each group, 4 animals were sacrificed at 4 and 8 weeks. The animals underwent randomized bilateral carotid artery injection with micropaque barium suspension just prior to sacrifice to help maintain vascularity. At harvest the implants and surrounding muscle and cranium were removed en bloc. New bone formation in the HA implants was examined by using routine histology and scanning electron microscopic backscattering image (quantitative) analysis. Microradiographs were performed on representative specimens. At 4 weeks postimplantation, backscattering analysis in the subperiosteal HA/BMP showed a mean 17.1% bone ingrowth vs. 11.3% of HA alone (p < 0.05). Supraperiosteal HA/BMP showed a mean 12.9% bone ingrowth vs. 0% of HA alone (p < 0.05). At 8 weeks, backscatter analysis of supraperiosteal HA/BMP showed a mean 19.33% bone ingrowth vs. 0% of HA alone (p < 0.05). Subperiosteal HA/BMP showed a mean 22% bone ingrowth vs. 20.85% of HA alone. This was the only group that did not have statistically significant results. Implant histology demonstrated woven bone within the interstices of HA/BMP placed either supra- or subperiosteally. In the HA-alone implants placed supraperiosteally, fibrovascular ingrowth was seen without any evidence of bone formation. In the HA-alone implant placed subperiosteally, woven bone was seen at the calvarium-implant junction. Microradiographs also demonstrated vascularization and bone formation similar to that seen on scanning electron microscopy. BMP-treated specimens appeared to have slightly greater vascularity than the nontreated specimens. The greatest bone formation occurred with the HA/BMP implant placed subperiosteally in the immature rabbit. Furthermore, these results demonstrate the potential prefabrication of vascularized bone flaps as early as 4 to 8 weeks. The clinical advantage of HA permits the surgeon to design osseous flaps that are customized in shape, fill all contour defects, and have little resorptive properties. Such prefabricated bone with an axial blood supply may allow for ultimate transfer as a pedicle or free flap to reconstruct osseous defects in children.