Bone Graft Materials

Every bone graft material the implant clinician will ever reach for can be judged by the same three questions, and almost all of the confusion in this field dissolves once those questions are asked in order. Does the material itself contain living bone-forming cells? Does it carry signals that recruit the host's cells and instruct them to make bone? And does it provide a physical scaffold that bone can grow along? These three properties — osteogenic, osteoinductive, and osteoconductive — are not interchangeable synonyms for "good graft." They are distinct biological mechanisms, and a given material may possess one, two, or all three.¹

The definitions are worth stating with precision because examiners and manufacturers alike blur them. A material is osteogenic when the graft itself contains viable osteoblasts and osteoprogenitor cells that directly lay down new bone. It is osteoinductive when it supplies biochemical signals — most famously the bone morphogenetic proteins (BMPs) and other matrix proteins — that recruit the host's mesenchymal cells and drive them to differentiate into osteoblasts. It is osteoconductive when it serves as a passive, biocompatible scaffold along which host bone and vessels grow, without contributing cells or signals of its own. Only one material in routine use delivers all three at once: autograft, the patient's own bone, which remains the gold standard precisely because it is simultaneously osteogenic, osteoinductive, and osteoconductive.¹³

The remaining categories trade some of that biological completeness for practical advantages — abundant supply, no second surgical site, or long-term volume stability. Allograft (processed human cadaveric bone, supplied as mineralized FDBA or demineralized DFDBA) is osteoinductive and/or osteoconductive but never osteogenic, because processing destroys all living cells. Xenograft (typically deproteinized bovine bone mineral) is osteoconductive only, but resorbs so slowly that it excels at holding contour. Alloplast (synthetic ceramics and glasses — β-TCP, hydroxyapatite, bioactive glass) is also osteoconductive only, with the advantage of unlimited supply and no disease-transmission risk. A recurring theme threads through all of them: the more biologically active a graft, the faster and less predictably it tends to resorb.¹

Because the three properties are independent, the cleanest way to internalize the whole field is to place each graft type on a map of which properties it carries. Autograft occupies the centre — the only material that sits inside all three domains. Demineralized allograft (DFDBA) adds inductive signalling to a conductive scaffold but brings no cells. Mineralized allograft (FDBA), xenograft, and the alloplasts all live in the purely conductive ring. The figure below renders this as a Venn diagram, which is the single most useful mental model in this chapter: it shows at a glance why autograft is irreplaceable in concept, and why every alternative is a deliberate compromise.

Autograft is bone harvested from the patient — intraorally from the ramus or chin, or extraorally from sites such as the iliac crest. It is the only material that carries living osteoblasts and progenitor cells (osteogenic), the full complement of matrix BMPs and growth factors (osteoinductive), and an autologous collagen–mineral scaffold (osteoconductive). Its costs are equally distinctive: a second surgical site with attendant donor-site morbidity, a finite available volume, and a resorption profile that can be rapid and somewhat unpredictable — which is precisely why autograft is so often combined with a slow-resorbing material to lock in volume.¹³

Allograft is processed human bone obtained from regulated tissue banks and supplied in two forms that behave very differently. FDBA (freeze-dried bone allograft) retains its mineral phase and functions primarily as an osteoconductive scaffold offering useful volume support. DFDBA (demineralized freeze-dried bone allograft) is treated with acid to remove the mineral, which exposes the matrix BMPs and confers osteoinductive potential.¹ The mechanistic subtlety here is examinable: FDBA contains the same BMPs in its organic matrix, but they remain trapped within the mineral and are unavailable to host cells until demineralization releases them — which is why mineralized FDBA is not reliably inductive despite carrying the proteins. Crucially, no allograft is osteogenic; processing and sterilization kill every cell. DFDBA's inductive potency, moreover, varies meaningfully between donors and processing batches.¹

Xenograft is animal-derived mineral, most commonly deproteinized bovine bone mineral (DBBM), processed to strip away the organic and immunogenic components and leave a mineral lattice structurally similar to human bone. It is osteoconductive only — a passive scaffold — but its defining clinical feature is slow, very low resorption. It persists for years, giving excellent long-term volume stability, which is why it dominates indications where holding contour matters: ridge preservation and sinus floor augmentation. Long-term (12–14-year) data on guided bone regeneration using DBBM with a barrier membrane support durable, stable peri-implant bone over time.²

Alloplast is fully synthetic — ceramics and glasses with unlimited supply and no donor site or disease-transmission risk, all osteoconductive only. The clinically important distinction is among the materials' resorption rates. β-tricalcium phosphate (β-TCP) resorbs relatively fast and is designed to be replaced as host bone forms. Hydroxyapatite (HA) resorbs slowly or is near non-resorbable, serving as a durable scaffold and long-term volume filler. Bioactive glass bonds chemically to bone and carries antibacterial properties. The resorption behaviour of an alloplast graft is therefore something the clinician selects deliberately by choosing the material or the ratio in a blend.¹

Material selection is, at root, a negotiation between the two axes developed above: how much biological activity the defect demands, and how much volume stability the result must retain over time. Where rapid new bone formation is paramount and the defect is well-contained and well-vascularized, lean on the active materials — autograft and the inductive DFDBA. Where maintaining contour over months to years dominates the goal — ridge preservation after extraction, or the sinus floor, where a fully resorbed graft would surrender the very space it was placed to create — choose a slow-resorbing scaffold such as DBBM or HA. And very often the right answer is to combine the two. Defect morphology, the presence of containing walls, host healing capacity, and the regulatory framework governing tissue-bank and animal-derived products all further shape the decision, which is why no single material is "best" in the abstract. The downstream surgical applications of these materials are developed in the augmentation chapters cross-referenced below.³