The role of insulin during the formation of bone in the augmented space of the maxillary sinus in patients with diabetes is unclear. The authors compared the differences in bone formation after maxillary sinus floor elevation in diabetic and healthy animals and evaluated the effects of insulin on osteogenesis and the differentiation and activities of the osteoblasts. 10 male Japanese white rabbits were divided into two groups after diabetic induction by a single injection of monohydrated alloxan and having maintained steady blood glucose levels. The groups included the diabetes mellitus group (DM; n = 5) and the DM + insulin group ( n = 5); another five healthy rabbits comprised the control group. Maxillary sinus floor elevation was performed by grafting hydroxyapatite particles. Compared with the control group, the newly formed bone area, number of blood vessels and osteoblasts, collagen I content and serum osteocalcin levels were significantly decreased in DM rabbits ( P < 0.01). Insulin treatment reversed the decrease in bone formation, blood vessels, osteoblasts, collagen I and serum osteocalcin ( P < 0.01). Insulin treatment also promoted osteogenesis in the augmented space of the diabetic rabbits, which might have resulted from promotion of osteoblast differentiation and upregulation of neovascularization.
The maxillary sinus graft procedure, also referred to as maxillary sinus floor elevation, is a well-established technique for increasing bone volume in the posterior edentulous maxilla before dental implant insertion . This technique has been performed by increasing numbers of implant surgeons because of its effectiveness and relatively low risk, but it is now known that uncontrolled diabetes is a contraindication for dental implants . Although the relationship between diabetes and dental implants has been reviewed , relatively little is known about the influence of diabetes on the success rate of maxillary sinus floor elevation. Studies using animal models have shown that diabetes leads to reduced bone formation, which results in osteopenia and delayed fracture healing . Previous studies exploring the influence of diabetes mellitus (DM) on bone formation have adopted models of experimental fracture of the tibia or femur and found that DM delays fracture healing and that insulin can reverse this effect . Bone formation in the limb bones is mainly through endochondral ossification, whilst bone formation in the jawbone is mainly through intramembranous ossification. The changes in bone formation in the augmented space of the maxillary sinus in patients with diabetes remain unknown.
Intramembranous ossification is characterised by direct osteoblastic differentiation without cartilage formation. In the present study, the authors investigated the number of active osteoblasts and their functional activities, including serum osteocalcin levels and collagen I content.
To produce a type I diabetic model, the authors intravenously injected monohydrated alloxan to impair the rabbits’ pancreatic β cells. In this model, maxillary sinus grafting was performed with and without insulin treatments. The aim of the present study was to compare the differences in bone formation after maxillary sinus floor elevation in diabetic and healthy animals and to evaluate the effects of insulin on osteogenesis and the differentiation and activities of the osteoblasts.
Materials and methods
All experimental procedures were approved by the Institutional Animal Care and Use Committee of Harbin Medical University, China. 6-month-old male Japanese white rabbits (body weight 2.5–3.0 kg) were obtained from the Animal Centre of the Affiliated Second Hospital, Harbin Medical University, China, and were bred in the animal facilities at Harbin Medical University.
Monohydrated alloxan, pentobarbital sodium, and insulin were obtained from Sigma (St. Louis, MO, USA). Mouse anti-rabbit monoclonal antibodies to osteocalcin were obtained from Abcam (Hong Kong, China). Hydroxyapatite ceramic particles were obtained from Beijing YHJ Science and Trade Co. (Beijing, China). Collagen membranes (Bio-Gide) were obtained from Geistlich (Wolhusen, Switzerland).
Induction of the rabbit diabetic model and experimental design
All the animals ( n = 21) were kept for 2 weeks to adapt to their environment. Blood samples were collected from the auricular veins of the animals for the evaluation of basal serum glucose levels using the glucose-oxidase enzymatic method. Five rabbits comprised the control group and the other 16 rabbits were subjected to the diabetic model.
In the experimental groups, diabetes was induced by a single intravenous injection of 150 mg/kg of body weight monohydrated alloxan (Sigma, St. Louis, MO, USA) dissolved in sterile 0.9% saline. Within 12 h of alloxan administration, a 5% glucose solution was offered to the animals to prevent hypoglycemia. At two time points, 72 h and 3 weeks after alloxan administration, the serum glucose levels were determined. Animals with serum glucose levels at or above 16.67 mmol/L at both time points were considered diabetic, and those with serum glucose level below 16.67 mmol/L were excluded from the study. Diabetes was successfully induced in 10 rabbits (from a total of 16) and they were randomly divided into two groups: the DM group ( n = 5) and the DM + insulin group ( n = 5).
After confirmation of the diabetic model 72 h and 3 weeks after alloxan administration, the rabbits underwent the surgical procedure. The hair was shaved and the surgical area was sterilized with iodophors and alcohol. The rabbits were anaesthetized by intravenous injection of pentobarbital sodium (1.5 mg/kg) dissolved in sterile 0.9% saline, and 0.5 ml of 1% lidocaine with epinephrine (1:100,000) was injected subcutaneously at the midline of the nasal dorsum for local anaesthesia. A 3 cm vertical midline incision was made as described by A sai et al. , and the skin and periosteum were elevated sufficiently to expose the nasal bone and nasoincisal suture line. Two circular nasal bone windows (diameter 4 mm) were made as shown in Fig. 1 A . The maxillary sinus membrane was kept intact during the procedure. The membrane was gently pushed inward and elevated from the floor, lateral wall, and medial wall of the maxillary sinus, and a compartment for graft materials was obtained . The hydroxyapatite ceramic particles (diameter 400–1000 μm; Beijing YHJ Science and Trade Co., Beijing, China) were mixed with the blood collected from the auricular vein of each rabbit, packed into the compartment, and slightly compressed ( Fig. 1 B). The bone windows were covered with a collagen membrane (Bio-Gide; Geistlich, Wolhusen, Switzerland) to prevent fibrous connective tissue ingrowth. The skin was then tightly sutured. A total of 200,000 U/day of penicillin was injected intramuscularly for 1 week after the operation. Figure 1 C shows the X-ray image of bone formation in the maxillary sinus after the rabbits were killed. The serum glucose levels were determined every week after the operation. Before the animals were killed, a blood sample was collected from the left ventricle for the last evaluation of serum glucose levels.
The DM + insulin group received twice-daily subcutaneous injections of swine insulin (10 U/day; Sigma, St. Louis, MO, USA). The rabbits from the control group and the DM group were injected with a sterile saline solution .
The animals were killed at the end of the eighth week after surgery. The rabbits were anaesthetized intravenously with pentobarbital sodium (Sigma, St. Louis, MO, USA). After 20 ml of blood was quickly collected from the left ventricle, the rabbits were euthanized with an overdose of pentobarbital sodium and perfused with a fixative containing a mixture of 5% glutaraldehyde and 4% formaldehyde through the left ventricle. The maxillary bones were dissected using a diamond saw. The right bone was kept in liquid nitrogen, and the left bone was placed in fixative for 48 h and demineralized with 10% EDTA for a week. The maxillary sinus portion was dissected and dehydrated in a graded series of ethanol and then embedded in paraffin. Serial sections representing the experimental site were cut in a frontal plane with a microtome set at 4 μm. From each specimen, nine sections representing the central part of the augmented area were selected for histological examination. Three sections were stained with haematoxylin–eosin (H–E), three with Sirius red, and three were stained using an immunohistochemical method for labelling osteoblasts.
Immunohistochemical method for labelling osteoblasts
Three sections from each specimen were immunostained using the streptavidin–biotin–peroxidase complex method. These sections were incubated with goat serum for 30 min and then incubated with a mouse anti-rabbit monoclonal antibody to osteocalcin for 30 min, rinsed in phosphate buffered saline (PBS) solution for 3 min, incubated for 30 min with biotinylated goat antimouse Ig, rinsed again in PBS, and finally incubated for 30 min with peroxidase-conjugated streptavidin. These sections were then rinsed in PBS again. Colour was visualized after a 5 min exposure to 3,3-diamino-benzidine tetrahydrochloride (DAB). Sections were rinsed in distilled water and counterstained with haematoxylin.
Histological examination and histomorphometric analysis
Histological examinations were performed using a Nikon 501 microscope (Nikon, Tokyo, Japan) and an Olympus polarized light microscope (Olympus, Tokyo, Japan) with an imaging software system (Image-Pro Plus 6.0; Media Cybernetics, Bethesda, FL, USA). For each animal, a total tissue area of approximately 9.8 mm 2 of the section stained with H–E was divided into three 40× fields (3.506206 mm 2 per field). The three fields for each animal were photographed, and the areas of newly formed bone were determined by drawing boundaries using the imaging software to calculate the percent mean value of the newly formed bone area.
A total area of approximately 9.8 mm 2 of the section stained with H–E was divided into 20 fields (0.49 mm 2 per field), and the area of highest vascularization at 200× magnification (i.e. 20× objective lens and 10× ocular lens; 0.14 mm 2 per field) was identified and photographed in each field (0.49 mm 2 per field). The numbers of blood vessel were counted, and the mean value for each animal was calculated. Any vessel lumen enclosed by standard endothelial cells (with or without red blood cells) that was clearly separate from adjacent blood vessels and other tissues (or with otherwise clearly defined borders) was considered a single, countable blood vessel. All the data were collected by two pathologists, and both had to agree on what constituted a single blood vessel before any vessel was included in the count.
Under the polarized light microscope, a total area of approximately 9.8 mm 2 of the section stained with Sirius red was divided into 20 fields (0.49 mm 2 per field), one image at 400× magnification (0.035 mm 2 per field) of each field (0.49 mm 2 per field) representative of bone formation was photographed. The total field area and collagen I area (stained red) were obtained by drawing boundaries using the imaging software to calculate the percent mean value of the collagen I area for each animal.
For immunohistochemical analysis, a total area of approximately 9.8 mm 2 was divided into 20 fields (0.49 mm 2 per field), and one image at 400× magnification (0.035 mm 2 per field) of each field (0.49 mm 2 per field) representative of bone formation was selected and photographed.
The osteoblasts with brown-stained cytoplasm were counted, and the mean number of osteoblasts per millimetre was calculated. Any brown cell on the surface of newly formed trabeculae that was clearly separate from adjacent cells or had a clearly defined border was considered a single, countable osteoblast.
All image analyses were performed by two pathologists who did not know the experimental design and were blinded to the identity of the specimens.
After centrifugation of the blood samples for 5 min at 3000 rpm, serum samples were obtained and stored at −80 °C for subsequent analysis. Serum glucose was measured using a FUJI DRI-CHEM 3500 (FujiFilm, Tokyo, Japan). Serum osteocalcin levels were measured using a radioimmunoassay kit (GenWayBioHeal Company, Beijing, China) and a γ radio-immunity counter (GC-γ radio-immunity counter; Zhongjiaguandian, China).
Data are presented as the mean ± SEM. Significance was determined by way of one-way ANOVA using SigmaStat analysis software (Systat Software, Chicago, IL, USA). A P value <0.05 was considered significant.
Serum glucose levels
Before maxillary sinus grafting, serum glucose levels were examined at 72 h and 3 weeks after alloxan injection. After the confirmation of the diabetic model, the surgical procedure was performed and the DM + insulin group was treated with insulin. Subsequently, serum glucose was examined once every week. As shown in Fig. 2 , the serum glucose levels in the DM group were significantly higher than those in the control group during the 8-week experiment period. Insulin treatment significantly reduced the elevated serum glucose levels in the DM + insulin group. These data indicated that the experimental conditions were reliable.
Newly formed bone
The newly formed trabeculae, hydroxyapatite particles, and fibrous connective tissues were seen in the augmented space ( Fig. 3 A) . In the control and DM + insulin groups, newly formed trabeculae were found in the interspace of the hydroxyapatite particles, immediately adjacent to the particles, or embedded in the fibrous connective tissues. The new trabeculae formed certain interconnections and networks and there were many vascular structures. No gap was found amongst the newly formed trabeculae, the hydroxyapatite particles, and the fibrous connective tissues ( Fig. 3 Aa and c). Limited numbers of trabeculae were formed in the DM group ( Fig. 3 Ab), and most of the augmented space was occupied by densely packed connective tissues and cells. The percentage of newly formed bone area was determined by dividing the newly formed bone area by the total area, and the data are presented in Fig. 3 B. The proportions of newly formed bone area were 0.2351 ± 0.0297%, 0.1309 ± 0.0133%, 0.2311 ± 0.0298% in the control, DM, and DM + insulin groups, respectively. Several of the trabeculae formed the primary osteons enclosing vascular structures in the control and DM + insulin groups ( Fig. 4 A and C) . In the DM group, most of the augmented space was occupied by provisional matrix comprised of densely packed connective tissues and cells ( Fig. 4 B). As shown in Fig. 4 (D and F), osteoblasts with cubic nuclei were apparent on the surface of the trabeculae that faced the connective tissues in the control and DM + insulin groups, but no osteoblasts were found on the surface of the trabeculae facing the hydroxyapatite particles. In the DM group, only a few osteoblasts were found on the surface of the trabeculae ( Fig. 4 E).