Effect of Enamel Matrix Derivative (Emdogain) on Bone Defects in Rabbit Tibias

Enamel matrix derivative (Emdogain, Biora AB, Malmö, Sweden) has been successfully used in periodontal regenerative procedures as reported by several investigations.1–6 Both animal and human studies have shown that Emdogain may lead to the formation of periodontium.2,6 The use of Emdogain in periodontal regeneration has been suggested because of its biological origin and properties. The original work of Slavkin and Boyde7 and Slavkin8 showed that proteins from the Hertwig's epithelial root sheath, related to the formation of enamel, gave initiation to the formation of acellular root cementum. Several studies followed.9,10 A major evidence of the role of enamel-related proteins and development of cementum came from studies from Hammarstrom's group.11 These studies in nonhuman primates clearly demonstrated that enamel matrix proteins are involved not only in the formation but also in the regeneration of the acellular cementum. Other in vitro studies concluded that enamel matrix proteins enhanced proliferation of periodontal ligament cells and that there was an increased production of proteins from periodontal ligament cells.12,13 Emdogain has been suggested as an alternative to bone grafts and barrier membranes in the treatment of intrabony periodontal defects.1 The aim of our study was to assess the effect of Emdogain on bone healing.

Ten New Zealand rabbits were used in the study, upon approval of the Ethical Committee for Human and Animal studies at the School of Medicine, University of Chieti. The rabbits of about 2.5 kg were kept in cages in the same institution. The methods followed a previously described protocol.14,15 The animals were anesthetized with a dose of ketamine (Ketalar, Parke Davis Spa, Milan, Italy,) and xylazine (Rompum, Bayer AG, Leverkusen, Germany). The ketamine was used to the dose of 44 mg/kg and the xylazine to the dose of 6 to 8 mg/kg for kilogram of weight. A local injection of 1.8 mL of lidocaine without vasoconstrictor was performed (Lidocaine, Astra, Södertälje, Sweden). A full thickness incision was performed to expose the upper anterior portion of the tibia. An 8-mm defect was created under sterile conditions, 1 per tibia, for a total of 2 defects per rabbit. The defects on the right legs were filled with Emdogain until the material was almost extruding from the defects. The left-leg defects were left unfilled (control). The surgical wounds were sutured with stainless steel monofilament wire 3.0 (Ethicon, Johnson and Johnson, Somerville, NJ). After the surgical procedure, a single dose of antibiotic was performed (0.25 g cefazolin intramuscularly). The postoperative course was uneventful. Five rabbits were killed with an overdose of Tanax T-61 after 4 weeks, and the remaining 5 were killed by the same method after 8 weeks. The area of interest in the tibia was exposed, and a block section was retrieved by means of a Stryker oscillating orthopedic saw (Stryker Instruments Co, Kalomazoo, Mich). The specimens were immediately fixed in 10% formalin and processed to obtain thin ground sections with the Precise 1 Automated System (Assing, Rome, Italy). For each specimen, 3 slides were obtained and stained with acid and basic fuchsin and toluidine blue. The slides were examined by a blind examiner in normal transmitted light under a Leitz Laborlux microscope (Leitz, Wetzlar, Germany).

The differences in the percentage of bone regeneration (new bone) between test (bone defect filled with enamel matrix derivative) and control (bone defect without enamel matrix derivative) sites were evaluated. The bone regeneration was expressed as the means ± SD and SE. The differences between the 2 groups were analyzed by the analysis of variance, and the Fisher PLSD and Scheffe F-test evaluated statistical significance of multiple comparisons. Level of significance was set at P≤ .05.

The control specimens showed bone healing with mineralized woven bone extending as trabeculae across the experimental site (Figure 1). The spaces between the bone trabeculae contained fibrous tissue and not the infiltration of lymphocytes and megacaryocytes typical of rabbit hematopoietic tissues. New osteoid matrix was present at the periphery of the bone defects, and many osteoblasts were observed (Figure 2). No trabecular bone was present in the central portion of the bone defects. The bone regenerated in the cortical portion was 30.4% ± 4.98%.

The initial formation of immature bone extending from the periphery of the bone cavities was evident (Figure 3). The rest of the bone cavity contained fibrous tissue with a mild inflammatory reaction. The inflammatory reaction was characterized by foci of lymphocytes distributed around the periphery of the cavity (Figure 4). The bone defect was not completely healed, and no remnants of Emdogain were visible. New bone extended also in the central part of the bone defects. The bone regenerated in the cortical portion was 30.2% ± 2.28%.

Mature and cortical bone was present in the cortical area, and small trabeculae of bone were present in the central zone of the tibia (Figure 5). Few small osteoblasts were observed (Figure 6). A small quantity of bone extended in the central part of the created defects. The bone regenerated in the cortical portion was 41.8% ± 1.31%.

Few small osteoblasts were present. Mature bone formed from the endosteal surface and filled the external portion of the bone defects (Figure 7). New bone and osteoid matrix were present in the central region of the defect. The periphery and central portion of the experimental cavity showed mineralized new bone formation. No enamel matrix was observed in the bone defects. Mature cortical bone was observed. Interestingly, in none of the groups were inflammatory cells or multinucleated cells present at the 8-weeks observation time. Formation, remodeling, and maturation of the bone tissue within the bone defects continued in the 8-weeks specimens (Figure 8). The bone regenerated in the cortical portion was 39.8% ± 6.31%.

At 4 weeks, statistical analysis of the difference between the test and the control bone defects was not significant (P= .860). Also at 8 weeks, the difference was not statistically significant (P= .507) (Table 1).

Emdogain has been recently introduced on the market for the treatment of periodontal defects as a biological agent that would induce periodontal tissues regeneration. Emdogain is obtained from porcine embryonic tissues and derives from amelogenins.16 Biologically, enamel matrix derivatives are effective on periodontal ligament cells as shown both by in vitro and in vivo studies. Embryologically, the enamel matrix derivatives derive from the Hertwig's epithelial sheet from which they are secreted. According to Hirooka,17 they play a major role in cementogenesis as well as in the development of the attachment apparatus. This aspect was, in fact, demonstrated by studies on monkeys from Hammarstrom's group.10 Studies from Petinaki et al18 have shown that Emdogain induces a light immunological response at concentrations far higher than those used in clinical practice. The same authors concluded that Emdogain can be safely used in clinical practice. In addition to efficacy, the clinical safety of the product has been shown in a study by Zetterstrom et al.19 These results confirm previously published data obtained from both animal and human histological specimens, as shown by the studies of Hammarstrom10 and Mellonig.6 We have used the rabbit tibia model in our study. The rabbit tibia model has already been described in this field by others,20,21 indicating the reliability and the biological relevance of the test. The results of our study indicate that Emdogain does not adversely affect bone formation and regeneration in induced bone defects. Paine and Snead22 have shown that amelogenin is directly involved in the biomineralization of the extracellular matrix in tooth formation. This indicates that amelogenin is involved in the mineralization of the tissue that is present. In fact, the studies by Paine and Snead show that amelogenin induces biomineralization of the extracellular matrix. This being true, it is not surprising that our study demonstrated that bone defects do heal in the presence of Emdogain. Histologically, no differences were present between test and control sites. In the 4-weeks specimens, it was possible to observe newly formed bone and a few osteoblasts, but it was not possible to distinguish the Emdogain. Mature bone was present after 8 weeks. In all specimens, no inflammatory or multinucleated cells were present. Our results showed that Emdogain is completely resorbed after a 4-week period and does not increase the formation of the new bone.

This work was partially supported by the National Research Council, Rome, Italy; and by the Ministry of Education, University, and Research, Rome, Italy. A special acknowledgment and thanks to Sig Marcello Piccirilli and Dr Giovanna Iezzi, DDS, for their assistance with the histological data.