A tissue-engineered bone biomimetic device was developed to regenerate calvaria critical-sized defects (CSDs) in athymic rats. Well-documented evidence clearly confirms that left untreated, CSDs will not spontaneously regenerate bone. To accomplish regeneration, four candidate treatments were assessed: porous poly(D,L-lactide) and type I collagen (PLC), PLC and human osteoblast precursor cells (OPCs) at 2 x 10(5) (PLC/OPCs), PLC and 50 microg of recombinant human bone morphogenetic protein-2 (PLC/rhBMP-2), and PLC/OPCs/rhBMP-2 (the bone biomimetic device). The hypotheses for this study were PLC/OPCs/rhBMP-2 would promote more new bone formation in CSDs than the other treatments and the amount of bone formation would be time dependent. To test the hypotheses, outcomes from treatments were measured at 2 and 4 weeks postoperatively by radiomorphometry for percent radiopacity and by histomorphometry for square millimeters of new bone formation. Data were analyzed by analysis of variance and Fisher's protected least significant difference for multiple comparisons with p< or = 0.05. At 2 and 4 weeks, radiomorphometric data revealed PLC/rhBMP-2 and PLC/OPCs/rhBMP-2 promoted significantly more radiopacity than either PLC or PLC/OPCs. Histomorphometry data at 2 and 4 weeks indicated significantly more new bone formation for PLC/rhBMP-2, PLC/OPCs/rhBMP-2, and PLC/OPCs compared to PLC. By 4 weeks, PLC/OPCs/rhBMP-2 and PLC/rhBMP-2 had regenerated the CSDs with more new bone than the other treatments; the quantity of bone at 4 weeks for these treatments was greater than at 2 weeks.