The osteoinductive potential of demineralized freeze-dried bone allograft in human non-orthotopic sites: A pilot study

Background: The use of demineralized freeze-dried bone allograft (DFDBA) in periodontal therapy is widely accepted. In addition to histologic studies demonstrating its role in regenerating the attachment apparatus, its recognized ability to exhibit osteoinduction in small laboratory mammals has prompted many clinicians to regard it as osteoinductive when utilized in periodontal therapy. Despite such a theoretical deduction, evaluating and establishing the osteoinductive potential of DFDBA in humans has been problematic. The purpose of this study was to investigate, through the use of a novel implantation/recovery model, the potential of DFDBA to induce new bone formation at sites not normally considered capable of de novo osteogenic activity. Methods: Seven patients scheduled to undergo periodontal therapy utilizing non-absorbable membranes agreed to have sterile polypropylene mesh capsules placed adjacent to the primary surgical site for the purpose of this investigation. One capsule was left empty while the second capsule contained DFDBA. At the appropriate time interval for removal of the therapeutic membrane, the capsules were removed and submitted for histologic analysis. Five of the subjects had procedures directed at regenerating bone within molar furcations (guided tissue regeneration) and had the membranes and associated capsules removed between 6 and 8 weeks. The 2 remaining patients had procedures directed toward ridge augmentation (guided bone regeneration) and had the membranes removed either at 8 or 9 months, respectively. In addition, one of the 2 long-term membrane patients had liberal cortical penetration performed directly beneath the implanted capsule. Results: Histologic analysis of all recovered capsules by 2 independent oral and maxillofacial pathologists could not confirm the presence of either osteoblastic or osteoclastic activity associated with the DFDBA particles, although the 2 longer-term specimens clearly exhibited trace amounts of vital bone non-contiguous with the implanted material. Conclusions: The results of this pilot study do not support the osteoinductive potential of DFDBA when utilized in quantities normally associated with periodontal bone grafting, although they do support the use of this implantation/recovery model to study other preparations and mechanisms of bone formation.