Differential effects of bone graft substitutes on regeneration of bone marrow.

BACKGROUND: This study used a rat tibial marrow ablation model to test the hypothesis that bone remodeling within the medullary canal varies with bone graft materials of different chemical compositions and structural properties, impacting marrow cavity restoration. Bone graft materials were selected based on their relative resorption or degradation in vivo and their osteogenic properties. METHODS: Following ablation of the right tibial marrow in male Sabra-strain rats, materials were implanted in the proximal marrow cavity: poly-D,L-lactide-co-glycolide 75 : 25 (PLGA); coralline-hydroxyapatite (HA), calcium-sulfate (CaSO4), collagen-HA-tricalcium phosphate granules, anorganic bovine bone mineral, demineralized bone matrix (DBM), 45S5 Bioglass (BG), PLGA with BG 50 : 50, PLGA : BG 80 : 20, and PLGA and PLGA:BG 50 : 50 plus bone marrow (BM). Control tibias were ablated but received no implants. At 2 (endosteal bone healing), 4 (marrow cavity remodeling) and 8 weeks (marrow restoration), six to eight animals per group were euthanized and tibias processed for histomorphometry of proximal and distal medullary canals. RESULTS: Control tibias showed primary bone in proximal and distal medullary canals at 2 weeks, with trabeculae surrounded by cellular marrow. At 4 and 8 weeks, control trabeculae were thinned and marrow had more fat cells. In the treated tibias, trabecular bone volume (TBV) varied with time and was material specific. Most implants supported comparable TBV at 2 weeks. Sites with CaSO4 or DBM exhibited decreased TBV with time whereas trabecular bone was retained in proximal tibias containing other materials, closely juxtaposed to the implants. TBV did not always correlate directly with implant volume, but changes in BM volume were correlated inversely with TBV. Addition of BM increased marrow restoration in sites containing PLGA; however, BM reduced restoration of marrow when added to PLGA : BG. Although the presence of implants in the proximal tibia resulted in retention of trabecular bone, there was a time-dependent reduction in TBV in distal canals; the rate and extent of the distal TBV reduction were implant dependent. CONCLUSIONS: Thus, although many materials can support bone formation in the marrow cavity, bone quality, quantity, and physical relationship to the implant, and its rate of resorption differ in a material-dependent manner, resulting in differences in the restoration of marrow. CLINICAL RELEVANCE: Bone graft materials should be selected not only for their ability to support new bone formation but also for their impact on the remodeling phase of bone healing.