Comparative analysis of the in vivo kinetic properties of various bone substitutes filled into a peri-implant canine defect model

Abstract

Purpose

Deproteinized bovine bone or synthetic hydroxyapatite are 2 prevalent bone grafting materials used in the clinical treatment of peri-implant bone defects. However, the differences in bone formation among these materials remain unclear. This study evaluated osteogenesis kinetics in peri-implant defects using 2 types of deproteinized bovine bone (Bio-Oss^® and Bio-Oss/Collagen^®) and 2 types of synthetic hydroxyapatite (Apaceram-AX^® and Refit^®). We considered factors including newly generated bone volume; bone, osteoid, and material occupancy; and bone-to-implant contact.

Methods

A beagle model with a mandibular defect was created by extracting the bilateral mandibular third and fourth premolars. Simultaneously, an implant was inserted into the defect, and the space between the implant and the surrounding bone walls was filled with Bio-Oss, Bio-Oss/Collagen, Apaceram-AX, Refit, or autologous bone. Micro-computed tomography and histological analyses were conducted at 3 and 6 months postoperatively (Refit and autologous bone were not included at the 6-month time point due to their rapid absorption).

Results

All materials demonstrated excellent biocompatibility and osteoconductivity. At 3 months, Bio-Oss and Apaceram-AX exhibited significantly greater volumes of formation than the other materials, with Bio-Oss having a marginally higher amount. However, this outcome was reversed at 6 months, with no significant difference between the 2 materials at either time point. Apaceram-AX displayed notably slower bioresorption and the largest quantity of residual material at both time points. In contrast, Refit had significantly greater bioresorption, with complete resorption and rapid maturation involving cortical bone formation at the crest at 3 months, Refit demonstrated the highest mineralized tissue and osteoid occupancy after 3 months, albeit without statistical significance.

Conclusions

Overall, the materials demonstrated varying post-implantation behaviors in vivo. Thus, in a clinical setting, both the properties of these materials and the specific conditions of the defects needing reinforcement should be considered to identify the most suitable material.

Graphical Abstract

INTRODUCTION

Since Dr. Branemark’s successful clinical application of a dental implant system in 1965, dental implants have become one of the primary methods for long-term treatment of lost teeth or dentition [1]. However, several challenges to implant application persist. These include insufficient bone, which causes distress for both surgeons and patients as it tends to complicate the surgical procedure and prolong the treatment time [2]. In such cases, bone augmentation is necessary [3]. Bone augmentation is the process of adding new bone to an area with insufficient bone mass, ultimately aiming to provide effective support and stability for subsequent implant placement [4]. Numerous common clinical causes of bone defects necessitate bone augmentation, such as periodontal disease, resorption or atrophy of the alveolar bone due to tooth extraction, and congenital anatomical defects like a low maxillary sinus floor [5,6]. Generally, after the addition of bone-filling material to the bone defect area, the material will gradually be replaced by new bone. The quality and quantity of the new bone, as well as the rate of material resorption, depend on the material used [7].

Autologous bone remains the gold standard for bone grafting due to its osteogenic, osteoinductive, and osteoconductive properties [8,9,10]. However, there are also many drawbacks, such as the tendency for invasion at the bone grafting site and resorption over time [11]. As a result, numerous bone-replacement materials with various properties have been developed. Two of the most commonly used materials in clinical practice are bovine xenograft materials and synthetic biomaterials [12]. Although many studies have investigated these bone-filling materials, the focus has primarily been on their unique properties and production methods, with few comparative studies conducted in animals. Consequently, when clinicians need to use bone-filling materials for bone augmentation, the choice is primarily based on personal clinical experience [13,14,15].

In this study, we aimed to characterize 4 distinct granular and spongy bone-filling materials, specifically 2 xenograft materials (Bio-Oss^® and Bio-Oss/Collagen^®) and 2 novel synthetic biomaterials (Refit^® and Apaceram-AX^®), each of which have distinctive characteristics, by evaluating their kinetic properties when used as filling material following implant placement in the extraction sockets of dogs. The findings may offer more objective data for clinicians, helping them to make more accurate selections of these materials for various patients and treatments.

MATERIALS AND METHODS

Animals

Eleven 18-month-old male beagle dogs, weighing 10 kg on average, were used in this study. All dogs were housed in separate cages, maintained on a 12-hour light/dark cycle, with a controlled room temperature of 23°C and a relative humidity of 40%. The dogs were fed soft food twice daily and had unrestricted access to water. All animal procedures adhered to the Tokyo Medical and Dental University guidelines for the care and use of laboratory animals. The experimental plan was approved and overseen by the Animal Care and Use Committee of Tokyo Medical University (study approval number: A2021-164C).

Bone substitutes

In this study, 4 different bone substitutes were used in addition to autologous bone, which served as a control. These substitutes included deproteinized bovine bone mineral (Bio-Oss^®, Geistlich Pharma, Wolhusen, Switzerland), deproteinized bovine bone mineral with 10% collagen (Bio-Oss/Collagen^®, Geistlich Pharma), interconnected superporous hydroxyapatite (Apaceram-AX^®, HOYA Technosurgical Co., Tokyo, Japan), and hydroxyapatite/collagen nanocomposite (Refit^®, HOYA Technosurgical Co.). Detailed material information can be found in Figure 1. Briefly, Bio-Oss has a porosity of approximately 60% and features both micropores (<10 μm) and macropores (>100 μm). It is one of the most widely used bone substitutes worldwide. Apaceram-AX has a porosity of 85% and contains 3 types of pores: macropores (50–300 μm), interconnecting pores (50–100 μm) that run between the macropores, and micropores (0.5–10 μm). Apaceram-AX has been shown to enhance initial bone regeneration compared to other synthetic hydroxyapatite substitutes without micropores [16]. Refit is characterized by its unique nanostructure that mimics natural bone, with an ultra-high porosity of 95% and large pores ranging from 100 to 500 μm in size, which allow rapid absorption [17].

Figure 1. Detailed information on the 4 different bone substitutes.

Surgical procedures

General anesthesia was administered using an intramuscular injection of ketamine hydrochloride (2 mg/kg Veterinary Ketalar, Daiichi Sankyo Propharma Co., Tokyo, Japan), along with local infiltration anesthesia using 2% lidocaine HCl with epinephrine 1:80,000 (Xylocaine, Dentsply, Tokyo, Japan) throughout the surgery. Intrasulcular incisions were made in the bilateral third and fourth premolars of the mandible in each dog, and mucoperiosteal flaps were elevated. Subsequently, the third and fourth premolars were carefully extracted. The extraction sockets and surrounding bone were ground using a round dental bur to create a defect model with a mesio-distal diameter of 8 mm, a buccolingual diameter of 3 mm, and a depth of 5 mm (Figure 2A). The autologous bone generated from grinding the jawbone was collected and used as a control. Rinsing with stroke-physiological saline was performed as needed throughout the entire process. One dental implant (Φ 3.2 mm; length, 10 mm, FINESIA, Kyocera, Tokyo, Japan) was placed in the center of each cavity (Figure 2B). Bone defects were randomly divided into 5 groups, including the control group, based on the material used to fill the gap between the implant and bone (Figure 2C). Materials were soaked in blood prior to insertion. After the placement of the bone substitute, the flap was repositioned and sutured. Animals were euthanized at either 3 or 6 months post-surgery. The mandibles were block-resected for micro-computed tomography (CT) and tissue staining analysis. The region of interest (ROI) was determined as the sum of the 2 rectangles on the left and right sides of the implant (Figure 3). The length was 4.8 mm from the upper edge of the peri-implant alveolar bone down the long axis of the implant, and the width was 2.7 mm from the lateral edge of the implant, extending away and perpendicular to the implant surface.

Figure 2. Mandibular defect model. (A) Schema representing the model dimensions. (B, C) After tooth extraction, the implants were placed in the extraction socket and filled with bone substitutes or autologous bone.

Figure 3. Region of interest selection for radiographic and histological analysis.

Radiographic analysis

At 3 and 6 months postoperatively, the specimens were analyzed using micro-CT (X-ray CT System SMX100CT, Shimazu Corp., Kyoto, Japan) at a voltage of 90 kV and a current of 90 μA to determine the area of newly formed bone. The images of the ROI for each material were obtained from 3 locations per graft site: the midline, 1 mm medial to the midline, and 1 mm distal to the midline. The area of new bone formation and residual materials in the ROI was measured using TRI/3D-BON-FCS64 software (Ratoc System Engineering Co., Ltd., Tokyo, Japan), and the average was calculated.

Histological analysis

Specimens were submerged in 10% buffered formalin for 72 hours, followed by dehydration through a series of ethanol gradients ranging from 70% to 100%. After dehydration, the specimens were embedded in methyl methacrylate resin. Longitudinal sections containing both tissue and implant were obtained using a high-speed, water-cooled diamond saw (EXAKT cutting system, EXAKT 300 CP, EXAKT Technologies, Norderstedt, Germany) and stained with Villanueva Goldner staining. The stained sections were then examined under a light microscope.

Histomorphometric analysis

Histomorphometric measurements were acquired to determine the bone occupancy of mineralized bone, osseous bone, residual bone substitutes, and other tissue components. The area of regenerated bone in the defect was assessed using histomorphometry software (OsteoMeasure, OsteoMetrics, Inc., Decatur, GA, USA). Furthermore, the bone-to-implant contact (BIC) was calculated by dividing the total length of calcified bone adjacent to the implant surface in the ROI by 4.8 mm (the vertical length of the ROI).

Statistical analysis

One-way analysis of variance and the Tukey’s honestly significant difference test were performed to compare differences among the materials. All statistical analyses were carried out using GraphPad Prism 9 (GraphPad Software Inc., La Jolla, CA, USA). A P value of <0.05 was considered to indicate statistical significance.

RESULTS

Radiographic measurements

At 3 months postoperatively, micro-CT results revealed that Refit exhibited a similar tendency to autogenous bone when compared to other materials. Although bone resorption was observed in the upper alveolar crest, akin to autogenous bone, the bone defect was replaced by new bone and no material remained (Figure 4). For the Bio-Oss/Collagen, Bio-Oss, and Apaceram-AX groups, the materials seemed to be gradually replaced by newly formed bone from the lower part of the defect, whereas an abundance of material remained in the upper region. At 6 months, Bio-Oss/Collagen and Bio-Oss showed progressive bone remodeling and resorption at the upper region, similar to the autogenous bone and Refit after 3 months. Apaceram-AX seemed to be replaced more slowly than Bio-Oss/Collagen and Bio-Oss, with material remaining at the crest. Measurements of the average newly formed bone and residual material volume showed that Bio-Oss had the highest volume after the 3-month period, followed by Apaceram-AX, Bio-Oss/Collagen, and Refit, the latter of which was comparable to autologous bone. All materials except Refit exhibited significant differences compared to the control (Figure 5). This changed after 6 months postoperatively, with Apaceram-AX exhibiting the highest volume, followed by Bio-Oss and Bio-Oss/Collagen. However, significant differences were observed only between Bio-Oss/Collagen and Apaceram-AX (Figure 5, Table 1).

Figure 4. Radiographic results. Views of the grafted areas at postoperative 3 months (RF, BC, BO, AP, and AB) and 6 months (RF and AB were excluded due to early resorption of the materials).

AB: autologous bone, RF: Refit, BC: Bio-Oss/Collagen, BO: Bio-Oss, AP: Apaceram-AX.

Figure 5. Newly formed bone measurements. The mean area of the newly formed bone and residual material filled in the defect site was measured.

AB: autologous bone, RF: Refit, BC: Bio-Oss/Collagen, BO: Bio-Oss, AP: Apaceram-AX.

^*P<0.05; ^**P<0.01; ^***P<0.001.

Table 1. Radiographic and histomorphometric measurements.

Parameters		3 mo					6 mo
		AB	RF	BC	BO	AP	BC	BO	AP
Newly formed bone and residual material (mm2)
	AVG	10.21	11.60	17.40	25.48	23.63	13.46	16.64	21.71
	SD	1.51	1.15	3.44	0.55	3.29	0.72	3.22	5.33
	SE	0.68	0.51	1.72	0.32	1.64	0.42	1.86	3.08
	95% CI	1.74	1.32	4.78	1.01	4.57	1.33	5.91	9.79
Mineralized bone (%)
	AVG	26.74	28.78	27.18	28.09	26.15	20.69	31.54	22.37
	SD	8.44	2.97	3.87	3.81	11.53	1.85	10.11	8.92
	SE	3.77	1.33	1.94	2.20	5.77	1.07	5.84	5.15
	95% CI	9.70	3.41	5.38	6.99	16.01	3.40	18.58	16.39
Osteoid (%)
	AVG	0.93	1.29	0.92	0.52	0.83	0.35	0.29	0.18
	SD	0.39	0.40	0.62	0.30	0.75	0.16	0.19	0.13
	SE	0.17	0.18	0.31	0.17	0.38	0.09	0.11	0.07
	95% CI	0.45	0.45	0.87	0.54	1.04	0.29	0.35	0.23
Residual material (%)
	AVG	0.00	0.02	10.07	21.66	49.14	5.61	5.22	37.14
	SD	0.00	0.05	6.55	2.46	13.18	3.04	3.11	6.89
	SE	0.00	0.02	3.28	1.42	6.59	1.76	1.80	3.98
	95% CI	0.00	0.06	9.09	4.53	18.29	5.59	5.72	12.67
BIC (%)
	AVG	58.56	59.11	75.59	66.09	68.72	68.02	57.88	55.40
	SD	14.34	12.67	13.77	3.71	14.95	14.03	11.85	8.71
	SE	6.41	5.66	6.88	2.14	7.47	8.10	6.84	5.03
	95% CI	16.49	14.56	19.11	6.82	20.75	25.78	21.78	16.00

AB: autologous bone, RF: Refit, BC: Bio-Oss/Collagen, BO: Bio-Oss, AP: Apaceram-AX, AVG: average, SD: standard deviation, SE: standard error, CI: confidence interval, BIC: bone-to-implant contact.

Histological and histomorphometric measurements

Figure 6 illustrates the occupancy of each component (mineralized bone, osteoid bone, residual material) and BIC at 3 and 6 months postoperatively. At 3 months postoperatively, the Refit group exhibited the highest percentage of mineralized bone occupancy (28.8%), followed by Bio-Oss (28.1%), Bio-Oss/Collagen (27.2%), autologous bone (26.7%), and lastly, the Apaceram-AX group (26.1%). Refit also demonstrated the highest osteoid occupancy at 3 months (1.3%), followed by autologous bone (0.9%) and Bio-Oss/Collagen (0.9%). This active replacement by newly formed bone was further corroborated by the histological specimens (Figure 7). Refit displayed rapid maturation, with cortical bone formation at the crest observed at the 3-month time point. However, no significant differences were detected in either mineralized bone occupancy or osteoid occupancy at 3 or 6 months.

Figure 6. Histomorphological measurements. (A-C) Occupancy ratios of the filled defect were measured for mineralized bone, residual materials, and osteoid, respectively. (D) Bone-to-implant contact ratio.

RF: Refit, BC: Bio-Oss/Collagen, BO: Bio-Oss, AP: Apaceram-AX.

^**P<0.01; ^***P<0.001.

Figure 7. Histological views (Villanueva Goldner stain) of the grafted areas of tested materials. The mineralized bone is stained green and the osteoid is stained red.

A large number of unresorbed bone substitute particles were observed on both sides of the implant in the Apaceram-AX, Bio-Oss/Collagen, and Bio-Oss groups (Figure 7). The proportion of residual materials for Apaceram-AX was found to be quite high at both time points (49.1% at 3 months and 37.1% at 6 months) (Figure 6). In comparison to Bio-Oss, which had a greater volume at 3 months, the rate of material absorption was slower for Apaceram-AX. For these 2 materials, the newly formed bone seemed to gradually expand from the inferior side of the defect along the implanted device (Figure 7). After 6 months, the maturity of the formed bone was evident as the area of woven bone decreased and bone marrow formation became prominent. Additionally, further new bone formation was observed in the superior region. However, the Bio-Oss/Collagen group, characterized by a high proportion of osteoids, tended to exhibit weak bone formation in the superior region, with soft tissue invasion observed in all samples after 6 months (Figure 7). The BIC showed no significant differences among all groups (Figure 6). The group with the highest BIC at both 3 and 6 months postoperatively was Bio-Oss/Collagen (75.6% and 68.0%, respectively). In comparison to 3 months postoperatively, the BIC decreased to varying extents in all groups at 6 months postoperatively, with the largest decrease in the Apaceram-AX group (−13.3%), followed by Bio-Oss (−8.2%) and Bio-Oss/Collagen (−7.6%) (Figure 6, Table 1).

DISCUSSION

Before placing an implant, it is crucial to assess the jawbone’s morphology at the intended implant site. When bone augmentation is necessary, the selection of bone graft material can significantly influence the peri-implant environment, depending on the bone’s morphology and the timing of its application. In this study, we examined the clinical behavior of various bone graft materials in dogs that received implants in their extraction sockets.

CT and histological results indicated that the Refit group had a similar bone formation rate to the autogenous bone group, with the alveolar crest being corticalized and the marrow showing maturity. Histomorphological analysis also revealed that at 3 months postoperatively, the Refit group exhibited the fastest resorption (with the implanted material almost completely gone), followed by the Bio-Oss/Collagen, Bio-Oss, and Apaceram-AX groups (averages of 89.9%, 78.3%, and 50.9%, respectively). This trend was maintained at 6 months as well, with the Bio-Oss/Collagen and Bio-Oss groups (94.4% and 94.8%, respectively) being absorbed more rapidly compared to the Apaceram-AX group (72.9%). Based on previous studies, we hypothesize that there are several reasons for the different absorption rates of the above materials. The first factor is the porosity and pore size of the materials. It is well-established that the higher the porosity and larger the diameter of the pores of a material, the larger the specific surface area of the pores [3,4,12]. As a result, osteoclasts and macrophages are more likely to adhere, which promotes fluid exchange within the material and increases the rate of resorption [18,19]. Second, due to the faster degradation of collagen [20], it seems obvious that Refit, which consists of both high porosity nanoparticles and collagen, would be absorbed faster. For the same reason, Bio-Oss/Collagen is also expected to have a higher bioresorption rate than Bio-Oss because it has the same inorganic composition, except that it contains 10% collagen. In general, the absorption rate of synthetic hydroxyapatite is considered to be slow, and Bio-Oss is also reported to have an extended absorption time [21]. However, in our study, we observed that Bio-Oss was absorbed relatively quickly compared to Apaceram-AX. Although Apaceram-AX had a higher resorption rate than other synthetic hydroxyapatite bone-replacement materials due to its high porosity and porous structure, the faster resorption rate of Bio-Oss may be due to the different degradation mechanism of the inorganic components compared to Apaceram-AX. Moreover, the most likely explanation for the discrepancy between Refit and Apaceram-AX, products of the same company with different collagen content, is because the high availability (HA) contained in Apaceram-AX is sintered, whereas Refit is not [17]. Due to this characteristic, the specific surface area of the HA particle is <40 nm in the long side for Refit, which is less than that of Apaceram-AX. In addition, taking into account the 95% porosity, the volume of HA included in Refit, composed of 20% collagen, would be approximately 4% in total, resulting in high apparent solubility. Together, these factors are believed to have caused the large difference in the amount of both residual materials.

In addition to the resorption rate of bone grafts, clinicians also consider the space retention capacity of bone substitutes as an important indicator [22,23,24]. A previous study by Yamada and Egusa [12] found a negative correlation between bone filler bioresorption and space retention capacity. In line with this study, significant bone resorption at the alveolar crest was observed in the Refit and autogenous bone groups, which may have led to inferior performance in space maintenance. Other experimental studies have reported favorable space maintenance outcomes after socket filling with Apaceram-AX and Bio-Oss [17,25]. Due to the limited evaluation period in this study, it is possible that the Apaceram-AX and Bio-Oss samples are still undergoing bone formation. As Apaceram-AX and Bio-Oss were not assessed beyond 6 months in this study, the extent of vertical bone loss was not analyzed. However, although further research is needed, it is highly probable that the space maintenance capability of these groups would be maintained for more than 6 months. The results in the upper region of the ROI measured in this study may, to some extent, reflect the amount of bone resorption in the alveolar crest. According to the concept of bone-guided regeneration [26], the use of a barrier membrane can prevent soft tissue from entering the graft and help maintain the morphology of regenerated bone. It is speculated that there may have been reduced bone resorption at the alveolar crest when applying a barrier membrane. The reason we did not use a barrier membrane in this study is because we wanted to eliminate the influence of the membrane and compare the kinetics of the graft material itself.

The mineralized bone mass of each material group was also compared. Consistent with the findings from previous studies [17,27], a slight decrease in mineralized bone volume was observed in the Apaceram-AX and Bio-Oss/Collagen groups at 6 months postoperatively compared to 3 months postoperatively. Interestingly, the Bio-Oss group was the only one that did not experience a decrease in mineralized bone volume. Furthermore, the Bio-Oss group exhibited a marginally higher mineralized bone mass at 6 months postoperatively compared to 3 months and demonstrated greater mineralized bone mass than the other groups. One potential explanation for Bio-Oss’s ability to resist alveolar bone resorption is the presence of CO₃. As indicated in prior studies, carbonate-containing apatite may enhance osteoconductivity, suggesting that it could promote bone regeneration [28,29].

In this study, all 4 materials demonstrated comparable BICs. Theoretically, BIC is positively associated with implant stability and interfacial strength, which result from the physical contact between the implant and the surrounding alveolar bone [30,31,32,33,34,35]. The BIC values observed in this study ranged from 55% to 74% across all groups, aligning well with the 50%–80% BIC found in clinically successful canine model implants reported by Block et al. [36,37]. Consequently, it can be inferred that the bone filler obtained in this experiment will likely yield similar implant stability when utilized for peri-implant defect repair.

Therefore, based on the results of this experiment, it is recommended to give greater consideration to the use of Bio-Oss and Apaceram-AX, which have low bone resorption rates, when restoring bone defects that require maintained contour or spacing (e.g., mandibular alveolar crest preservation or maxillary sinus floor elevation). However, clinicians should be aware that the time required for these materials to degrade remains a subject of debate. Meanwhile, although Refit and Bio-Oss/Collagen—which have high resorption rates—may be disadvantageous in bone defects such as lateral bone defects that require space preservation, they may be suitable for 4-walled bone defects or extraction sockets, which are considered to have relatively favorable regenerative ability. In such cases, an early transition to implants may be planned. It is worth noting that, although Refit was the least space-maintaining among the materials compared in this study, the induced bone was equivalent to autologous bone, which is the gold standard. In this study, the grafting material was filled at approximately 120% of the defect model, but it may be possible to achieve better results by further overcorrecting the defects using Refit.

Differences in performance were observed among the materials tested. Our results indicate that no single bone-filling material consistently demonstrates superior properties compared to the others. However, there is still much to be clarified regarding these materials. For instance, for those materials exhibiting a slow absorption rate, how long will they persist? Future experiments should incorporate more qualitative evaluation methods and additional time points. Furthermore, although the defect model employed in this study simulates a 3-walled defect, the size and shape of the environment differ from the bone defects and extraction sockets encountered in actual clinical practice. Consequently, careful consideration is necessary when interpreting our findings and designing subsequent experiments.

In conclusion, we developed a peri-implant bone defect model to estimate bone kinetics in human peri-implant bone defects. The findings may assist clinicians in choosing appropriate bone-filling materials.