The authors have reported that a scaffold constructed of synthetic octacalcium phosphate (OCP) and porcine atelocollagen sponge (OCP/Col) enhanced bone regeneration more than sintered β-tricalcium phosphate collagen composite or sintered hydroxyapatite collagen composite with a rat calvarial defect model. To aim for clinical application, the present study investigated whether OCP/Col would enhance bone healing in a dog tooth extraction socket model. Six adult, male, beagle dogs were used. The tooth extraction socket model was made by extracting bilateral third maxillary incisors and the subsequent removal of buccal bone. Disks of OCP/Col were implanted into one side of the model and the other side was untreated. The specimens were fixed 1 or 3 months after implantation. In radiographic analysis, the OCP/Col-treated group showed a wider range of radiopacity than the untreated control. Histologically, the OCP/Col-treated group showed more abundant newly formed bone than untreated control, and the implanted OCP was gradually resorbed. In morphometrical analysis, enlargement of the buccal alveolus in the OCP/Col group was significantly greater than in the untreated control. This study showed that implanted OCP/Col would be replaced by newly formed bone and OCP/Col implantation would enhance bone healing in a tooth socket model.
Bone repair, such as reconstruction of bone defects after extirpation of tumours and cysts of the jaw, alveolar bone grafting for patients with cleft lip and palate, and maxillary sinus augmentation for implant surgery, is a major problem in maxillofacial surgery. It is generally accepted that autogeneous bone grafting is an effective technique for critical-sized defects, but it has disadvantages such as limited availability and morbidity associated with harvesting bone from a second operative site . Synthetic biomaterial has been developed as a bone substitute to overcome the disadvantages of autogeneous bone grafting, and hydroxyapatite (HA) and β-tricalcium phosphate (β-TCP) have been applied in clinical use . Autogeneous bone grafting is still the gold standard procedure, because substitute materials do not show superior bone regeneration to autogeneous bone grafting.
Octacalcium phosphate (Ca 8 H 2 (PO 4 ) 6 ·5H 2 O; OCP) is a calcium phosphate and has been suggested to be a precursor of biological apatite in bone . The osteoconductive characteristics of synthetic OCP were first established in the subperiosteal region of mouse calvaria . Previous studies by the authors show that synthetic OCP enhances bone regeneration more than HA and β-TCP, if implanted into a critical-sized defect of rat calvarium . Excellent osteogenicity is shown by OCP coating of metal implants . OCP cannot be molded by a sintering process, unlike bioceramics, such as HA or β-TCP, because of the intrinsic crystal structure of OCP, including a large number of water molecules within the layered HA structure .
To improve handling performance, the authors have developed a composite of synthetic OCP and porcine atelocollagen sponge (OCP/Col) . The synergistic effects of OCP/Col for bone regeneration are composed of bone nucleation by OCP and cell infiltration by collagen. OCP/Col significantly enhances bone regeneration more than OCP alone, collagen alone, β-TCP collagen composite (β-TCP/Col), and synthetic sintered ceramic HA collagen composite (HA/Col) without cell transplantation and exogenous osteogenic cytokines, if implanted into a critical-sized calvarial defect rat model . Recently, the authors succeeded in implanting OCP/Col into a critical-sized calvarial defect canine model to aim for clinical application, and confirm the efficacy of bone regeneration by OCP/Col .
The authors thought that OCP/Col implantation would enhance bone healing in a dog tooth extraction socket model, because a tooth extraction socket, which is the most common defect in oral and maxillofacial surgery, is usually a non-critical-sized defect. Several approaches have been used to prevent post-extraction ridge reductions , but it would be profitable for preprosthetic surgery if OCP/Col were applicable for bone regeneration of the tooth extraction socket. The present study investigates whether OCP/Col implantation would enhance bone healing in a dog tooth extraction socket model. The process of bone regeneration is compared morphometrically, radiographically and histologically between OCP/Col and an untreated control.
Materials and methods
Preparation of composite of OCP/collagen disks
OCP was prepared by mixing calcium and phosphate solution as described previously . The sieved granules (particle size 300–500 μm) of OCP, obtained from dried OCP, were sterilized by heating at 120 °C for 2 h. The authors’ previous study showed that such heating does not affect physical properties, such as the crystalline structure or specific surface area of OCP granules , although it was reported that increasing the temperature above 100 °C induced collapse of the OCP structure because of dehydration . Collagen was prepared from NMP collagen PS (Nippon Meat Packers, Tsukuba, Ibaraki, Japan), a lyophilized powder of pepsin-digested atelocollagen isolated from porcine dermis. OCP and collagen composite (OCP/Col) was prepared from NMP collagen PS and OCP granules . OCP was added to the concentrated collagen and mixed. The OCP/Col mixture was then lyophilized, and the disk was molded (9 mm diameter, 1 mm thick). The molded OCP/Col underwent dehydrothermal treatment (150 °C, 24 h) in a vacuum drying oven DP32 (Yamato Scientific, Tokyo, Japan) and were then sterilized using electron beam irradiation (5 kGy).
Eighteen-month-old, male, beagle dogs ( n = 6; NARC Co., Chiba, Japan) were used. The principles of laboratory animal care, and national laws, were followed. All procedures were approved by the Animal Research Committee of Tohoku University. General anaesthesia was administered with intravenous sodium pentobarbital (0.5 ml/kg), followed by intramuscular atropine sulfate (0.5 mg) and ketamine hydrochloride (20 mg/kg). After disinfection of the oral cavity and injection of local anaesthesia (2% lidocaine with 1/80,000 epinephrine), a tooth extraction socket model was made by extracting the bilateral third maxillary incisors and the subsequent removal of buccal bone to the extent of the root apex ( Fig. 1 ). Buccal marginal gingiva was incised from the maxillary second incisor to canine tooth, and a vertical incision was made on the distal side of the canine tooth, and then the muco-periosteum was ablated. After extraction of the maxillary third incisor, sufficient overlying bone on the buccal side was removed to avoid damaging the neighbouring tooth. Five–nine disks of OCP/Col were implanted into one side of the tooth extraction socket model, and the other side was untreated as a control. The tooth extraction socket model was closed and made watertight. To prevent infection, flomoxef sodium was administered by intravenous drip during the operation, and cefcapene pivoxil hydrochloride hydrate was used postsurgically by oral administration for 3 days. A soft diet was fed to rest the operative wounds for 2 weeks after implantation. Three dogs were killed 1 month after implantation and 3 dogs were killed 3 months after implantation by intravenous injection of an overdose of sodium pentobarbital. After death, the maxilla and surrounding tissues were resected and fixed with 10% formalin neutral buffer solution, pH 7.4, by infiltration for 4 weeks at 4 °C.
Morphometric analysis of study model
There were four experimental groups (OCP/Col 1 M, untreated 1 M, OCP/Col 3 M, untreated 3 M) each of which had three samples. Three-dimensional (3D) morphometric differences between OCP/Col 1 M and untreated 1 M, and between OCP/Col 3 M and untreated 3 M were analysed using a study model. To prepare the study model, an impression of the resected maxilla was taken immediately after death. The hypothetical standard plane of the study model was defined as the most apical points connecting the gingival margin at bilateral second incisors and bilateral canine teeth. A base plane of the study model was prepared 10 mm from the hypothetical standard plane (palatal analytic model; Fig. 2 a ). The palatal analytic model was plotted using a 3D scanning machine (MODELA Model MDX-15; Roland DG Corp., Hamamatsu, Shizuoka, Japan), and a 3D digital image was constructed using public domain software (Graph-R). The palatal analytic model was analysed by observing from the palatal side of the maxilla. To measure buccopalatal protrusion after the operation, two standard lines from the most distal point of the second incisor to the most buccal and palatal points of the canine in the oral cavity were drawn. The area enclosed with the two standard lines and the mesial aspect of canine tooth was defined as the buccopalatal triangle ( Fig. 2 c). The percentage of buccopalatal protrusion (BPP %) was calculated as the area of tissue in the buccopalatal triangle/area of buccopalatal triangle × 100. The BPP % was quantified on a computer using Scion Image public domain software (Scion Corporation, Frederick, MD).
After measuring BPP %, the palatal analytic model was trimmed frontally on the plane across the centre of incisal papilla, and it made a frontal analytic model ( Fig. 2 b) to measure a buccopalatal hollow after the operation. A perpendicular line to the base plane of the study model across the centre of incisal papilla was defined as the Y axis. The distance from the Y axis to the most hollow point (H-distance) of the OCP/Col (O) or the untreated control (C) was compared ( Fig. 2 d).
Histomorphometric data were analysed using a computer software package (Excel v.X; Microsoft Co., Redmond, WA, USA). All values were reported as the means ± standard deviation (SD). The χ 2 test was applied to examine whether each group had a normal distribution, the F -test examined whether the standard deviations of two normally distributed groups were equal. Paired t tests were used to compare mean values between OCP/Col 1 M and untreated 1 M, and between OCP/Col 3 M and untreated 3 M. Statistical significance was accepted at p < 0.05.
Radiographic analysis and tissue preparation
Maxillary occlusal radiographs were taken by dental radiographic apparatus (SANKO X-ray MFG) with instant film for occlusal radiography (Hanshin Technical Laboratory, Ltd., Nishinomiya, Hyogo, Japan) under standardized conditions (50 kV, 10 mA, 0.7–1.0 s) immediately after operation. After death, at 1 and 3 months postoperation, the resected maxillae were radiographed using a microradiography unit (Softex CMBW-2; Softex, Tokyo, Japan) with X-ray film (FR; Fuji Photo Film, Tokyo, Japan) under standardized conditions (40 kV, 5 mA, 60 s) in which neither OCP nor OCP/Col showed radiopacity. After the radiographs were taken, the samples were cut horizontally into two pieces at the centre of the defect.
One piece of each sample was dehydrated in a graded series of ethanol, embedded in methyl methacrylate and sectioned horizontally using a low-speed saw machine (Isomet 5000; Buehler, Lake Bluff, IL, USA) with a diamond wafering blade. The sectioned wafers were mounted on plastic slides and were ground and polished until they were 200–300 μm thick. Contact microradiography of undecalcified sections was taken with a microradiography unit (Softex CMR Unit; Softex, Tokyo, Japan) for 60 s exposure, at 20 kV, 5 mA.
The other pieces of each sample were decalcified in formic acid and sodium citrate solutions for 12 weeks at 4 °C. The samples were dehydrated in a graded series of ethanol, embedded in paraffin and sectioned horizontally at a thickness of 10 μm. The sections were stained with haematoxylin and eosin (HE), and photographs were taken with a photomicroscope (Leica DFC300 FX; Leica Microsystems Japan, Tokyo, Japan).
Fourier transform infrared spectroscopy of implanted OCP/Col
Samples from the OCP/Col implanted maxillae at 3 months were examined with Fourier transform infrared spectroscopy (FTIR). Samples were collected from one maxilla of a dog from the 3-month implantation group. Tweezers were used for collection to exclude as much soft tissue around the implanted OCP/Col as possible. The implants were immediately washed in deionized water, lyophilized and ground to powder. Samples were then collected for FTIR. The FTIR spectrum of each sample was obtained by a HORIBA FTIR FREEXACT-2 (HORIBA, Kyoto, Japan), with the sample diluted with KBr over a range of 2000–500 cm −1 with 4 cm −1 resolution. As controls, collagen, OCP, OCP/Col, HA and a dog bone were examined with FTIR to compare with implanted OCP/Col.
Morphometric analysis of study model
In both experimental periods for the excised maxilla, the OCP/Col-implanted group demonstrated a convex alveolus at the operated site, whereas the untreated group showed a distinct concaved alveolus at the operated site ( Fig. 3 a ). By inspecting the palatal analytic model, maxillary alveolus treated with OCP/Col demonstrated buccopalatal protrusion, whereas untreated maxillary alveolus became concave.
Each group (OCP/Col 1 M, untreated 1 M, OCP/Col 3 M, untreated 3 M) indicated normal distribution and each group comparison (OCP/Col 1 M vs. untreated 1 M, OCP/Col 3 M vs. untreated 3 M) showed equal distribution, so paired t tests were applied to compare mean values between OCP/Col 1 M and untreated 1 M, and between OCP/Col 3 M and untreated 3 M. The percentages of buccopalatal protrusion (BPP %) are shown in Fig. 3 b. BPP % ± SD in the OCP/Col 1 M and untreated 1 M was 64.1 ± 4.1 and 28.8 ± 6.9, respectively. A significant difference was seen between the OCP/Col 1 M and untreated 1 M (paired t test, F = 0.352; df = 4; P = 1.62 × 10 −3 ). BPP % ± SD in the OCP/Col 3 M and untreated 3 M was 74.3 ± 2.4 and 32.7 ± 1.6, respectively. A significant difference was seen between the OCP/Col 3 M and untreated 3 M (paired t test, F = 2.32; df = 4; P = 1.38 × 10 −5 ) ( Fig. 3 b). Morphometric analysis regarding the distance from the Y axis to the most hollow point (H-distance) is shown in Fig. 3 c. H-distance ± SD in the OCP/Col 1 M and untreated 1 M was 13.3 ± 0.38 mm and 11.2 ± 0.85 mm, respectively. A significant difference was seen between the OCP/Col 1 M and untreated 1 M (paired t test, F = 0.199; df = 4; P = 0.0179). H-distance ± SD in the OCP/Col 3 M and untreated 3 M was 12.4 ± 0.72 mm and 10.2 ± 0.94 mm, respectively. A significant difference was seen between the OCP/Col 3 M and untreated 3 M (paired t test, F = 0.588; df = 4; P = 0.0287) ( Fig. 3 c).
OCP/Col disks before implantation showed no radiopacity under standardized conditions, whereas sintered HA granules, as a control, showed granulous radiopacity. On maxillary occlusal radiographs taken immediately after operation, the OCP/Col-implanted area as well as the untreated site showed hardly any radiopacity, and both operated areas became concave ( Fig. 4 a ). One month after operation, the OCP/Col-implanted area showed a foggy convex radiopaque mass, whereas the untreated site showed distinct concave radiopacity comparable with the surrounding maxilla ( Fig. 4 b). Three months after operation, the OCP/Col-implanted area showed a condensed radiopaque mass and convexity persisted. The untreated site showed concave radiopacity ( Fig. 4 c).