DISCUSIÓN

Experimental study in rats of mandibular bone regenaration with different biomaterials

Estudio experimental sobre la regeneración ósea mandibular de la rata con diferentes biomateriales

Santiago Ochandiano Caicoya

Médico Adjunto. Servicio de Cirugía Oral y Maxilofacial. Hospital General Universitario Gregorio Marañón. Madrid. España

It is calculated that approximately one million bone grafts are made yearly in the United States, including bone grafts by traumatologists, neurosurgeons, maxillofacial surgeons, and odontologists. Innumerable studies have been published on bone regeneration in the course of searching for an ideal substitute for autologous bone. However, many studies repeat methodologic errors, draw conclusions that do not adhere strictly to the study findings, and, on a few occasions, are biased by commercial interests. The same biomaterial may have reports of very favorable results or clear regeneration failures. This is confusing for readers, who find it difficult to differentiate between valid and questionable scientific findings. Report often concluded finally with "I do it that way," I use it and it works," and statements to the effect. No real conclusions are drawn about whether a given biomaterial improves ossification.

To avoid renouncing that goal, we have to begin at the beginning, with experimental animal models. We would like to welcome the study by doctors B. Peral Cagigal, L.M. Redondo González, as an investigation that provides the necessary rigor for scientific studies. In the first place, they defined and demonstrated in an adult rat mandibular model a critically sized through-and-through bone defect, i.e., a defect incapable of regenerating spontaneously in the course of the animal’s life.¹ This is important because studies of non-critically sized defects do not allow us to draw conclusions about the biomaterial used. It also is very important not to study only linear filling of the defect but also three-dimensional regeneration, i.e., the bone volume obtained. This is a much more exact parameter for verifying that bone regeneration has occurred.²

As defined by Urist³ and Reddii,⁴ a material is an osteoinductor if it can generate bone in locations where it would not exist in normal conditions; bone is obtained by phenotypical conversion of connective tissue into normal bone. This refers to heterotopic bone formation, or bone outside the skeleton (e.g., BMPs form bone in the midst of muscle). In order to define a biomaterial as osteoinductive, this property has to be demonstrated in animal models. Therefore, osteoinduction cannot be defined by this study because bone regeneration is in an orthotopic location. Consequently, we do not share the authors’ claim, in the introduction of Lambone^® membrane, that: "This membrane has osteoinductive and osteoconductive capacity." The osteoinductive capacity of Colloss^® mentioned in the introduction also has to be proven in a suitable experimental model.

Osteoinduction, moreover, is characterized by the presence of ossification foci in the center of the defect, not by bone regeneration from the walls of the defect. The histologic preparations of the authors indicate that reossification of the defect took place starting from the limits of the bone defect and progressing toward the center of the defect. Consequently, only osteoconductive properties are involved.⁵

With respect to collagen type I, it is one of the two biomaterials studied that had the best osteoconductive result. Complete ossification of the defect from the margins of the defect was evident at six weeks, both radiologically and histologically. This indicates a clear advantage of this group over the control group or the membrane alone group.

An important conclusion reported here is that bioactive glass could be an obstacle for ossification if we compare the bone found in the membrane alone group with the membrane plus bioactive glass group. It is noteworthy that the known mechanism of replacement of bioactive glass is dissolution in the absence of a foreign body reaction of strange body. This means that replacement is mediated passively by macrophages or polymorphonuclear cells. However, the authors found "intense foreign-body inflammatory cellularity."

The authors’ conclusions with regard to this specific model of through-and-through circular critical defect are clear: the best regeneration was achieved using collagen type I and membrane, and the next best was obtained with membranes alone. The use of glass with a membrane slowed the reossification process. In this case, the biomaterial did not provide any added advantage and, in fact, impeded reossification in the first stages. As Schenk and Buser⁶ state, the advantages of covering defects with a membrane is: to prevent the collapse of the space where regeneration will occur, maintain a blood clot in place, which allows the migration of osteogenic cells and protects the delicate network of new vessels that develop inside the clot. Finally, a membrane keeps out nonosteocompetent cells that would try to create a fibrous scar in the area.

These conclusions do not translate automatically into clinical use, particularly when referring to osteoconduction rather than osteoinduction. Bone metabolism and bone replacement or regeneration in murine experimental models have little to do with human bone behavior.

In molecular biology studies in cancer, many treatments that are successful in mice fail when tested in humans. It is important in bone regeneration studies to seek experimental models animals that have a bone metabolism similar to humans, such as mammals like monkeys or, in our setting, pigs or dogs. The FDA, for example, requires demonstrating the effectiveness of an agent in a small-animal experimental model and then in a larger and heavier animal with demonstrated cortical remodeling before osteoporosis agents are accepted for clinical trial.⁷ It is known that the bone that has the composition most similar to human bone is canine bone; rat bone is the bone least similar to human bone. However, human bone really does not resemble any other animal in measurements of bone density and mechanical behavior. This may explain why humans have a much higher incidence of fractures than animals. Therefore, no ideal animal model of bone exists. While it is accepted that experimentation in rats is a good method for preliminary screening, studies must be completed in other species.⁸

Finally, I would like to congratulate the authors for their work, scientific rigor, and the interesting conclusions obtained.

References

1. Anderson ML, Dhert W, De Brujin JD, Dalmeijer RA, et al. Critical size defect in the goats os ilium. Clin Orthop Rel Res 1999;364:231-9.        [ Links ]

2. Gosain AK, Song L, y cols. Osteogenesis in cranial defects: reassessment of the concept of critical size defect and the expression of TGF-B isoforms. Plast Reconstr Surg 2000;106:360-371.        [ Links ]

3. Urist M. Bone: formation by osteoinduction 1965;150:893-9.        [ Links ]

4. Redii AH, Hugins CB. Biomechanical sequences in the transformation of normal fibroblasts in adolescents rats. Proc Nal Acad Sci. EE.UU. 1972;69:1601-5.        [ Links ]

5. Glowacki J. A review of osteoinductive testing methods and sterilization processes for demineralized bone. Cell and Tissue Banking 2005;6:3-12.        [ Links ]

6. Schenk R, Buser D, Hardwick R, Dahlin C. Healing Pattern of Bone Regeneration in Membrane-Protected Defects: A Histologic Study in the Canine Mandible. Int J Oral Maxillofac Impl 1994;13:9-29.        [ Links ]

7. Thompson DD, Simmons HA, Pirie CM, Ke HZ. 1995 FDA guidelines and animal models for osteoporosis. Bone 17:125S-133S.        [ Links ]

8. Aerssens J, Boonen S, Lowet G, Dequeker J. Interspecies differences in bone composition, density, and quality: potential implications for in vivo bone research. Endocrinology 1998;139:663-70.        [ Links ]