In orthopedic and oral and maxillofacial surgery, screws and plates are needed to fix the bones, provide space for guided bone regeneration (GBR), and maintain the bone shape when there are fractures or lesions. To maintain support strength and biocompatibility, screws and plates made of titanium and its alloys are used [1-5].

Although titanium and its alloys have biocompatibility; however, because they are metals, they are associated with various problems such as corrosion, subsequent elution of metal ions, and the inconvenience of removing metal screws and plates through a second surgery [6-8]. Therefore, it is necessary to develop an absorbable screw and plate to overcome these problems.

The ideal properties of absorbable graft materials are biocompatibility, bioactivity, and bioresorbability. Synthetic graft materials must degrade into nontoxic products through physiological mechanisms in a controlled manner to provide space for the ingrowth of new bone from the surrounding bone, which should completely replace the graft material [9].

β-tricalcium phosphate (β-TCP) and hydroxyapatite (HA) are the most commonly used calcium phosphates in ceramics [10-13]. HA has excellent biocompatibility and osteoconductivity. It has a chemical structure that closely resembles natural bone mineral. However, the biodegradation of HA is too low to achieve the optimal formation of bone tissue, and poor biodegradation prevents natural bone ingrowth for extended periods [14,15]. β-TCP is a bioactive and biodegradable graft material. It has the advantage of being easily absorbed in vivo, but it lacks the physical properties. Moreover, the bioresorption rate of β-TCP is unpredictable, thereby affecting its bone-forming ability [16].

To improve the defects of β-TCP and HA, a composite with these two materials in appropriate proportions was introduced [17-19]. A previous study reported that TCP–HA composites (biphasic calcium phosphate, BCP) showed better bioabsorbability than calcium phosphates alone [18,19]. Various BCP ratios for osseointegration were studied, such as 40:60, 70:30, 60:40, and 80:20 and recently, the 70:30 ratio was identified as the most appropriate.

Several studies examined whether screws and plates made of calcium phosphates can be used. Calcium phosphate screws and plates have been considered an effective fixation system with several advantages over metallic fixation; for example, there is no need to remove the materials after osseous healing, no worries for corrosion and accumulation of metal in tissues, and less pain and reduced stress-shielding because the materials initially bear less load and gradually transfer the load as they degrade [20,21]. However, there are some problems related to the use of calcium phosphate screws and plates such as, rapid loss of initial strength and weakness compared with metallic implants [22-26]. To overcome the disadvantages of calcium phosphate screws and plates, studies were conducted on adding polymer or TCP to HA [27-31].

Recently, studies have suggested that BCP can be used not only as bone-regenerating material but also as a substitute for metal materials in vivo [32,33]. If screws and plates made of such absorbent materials are used in bone reconstruction instead of metal screws and plates, it may be more beneficial. However, only a few studies have been conducted on screws and plates made of BCP. Some related studies have been conducted, but they have only evaluated their biocompatibility in experimental studies, and experiments were conducted on small animals such as rats and rabbits. In addition, no preclinical large-animal studies or clinical trials have been conducted on the use of BCP.

Therefore, this study aimed to evaluate the physical properties and biocompatibility of BCP and its clinical applicability by performing a histological analysis of BCP screws inserted into the mandible of beagles.

Sample preparation

TCP-HA composite (BCP) fabrication

TCP and HA were purchased from Berkeley Advanced Biomaterials Inc. (Berkeley, CA, USA). The powder particle size of TCP and HA was 100 nm. BCP used in this study was prepared at a ratio of 70:30. The experimental preform was prepared by compressing TCP, HA, and BCP. For cell experiments, a disk-shaped preform with a diameter of 15 mm and a height of 2 mm was made, and for animal experiments, a screw-type preform with a diameter of 1 mm and a length of 4 mm was made. Each preform was processed by heating it to 850°C, with a temperature increase rate of 8°C/min. All manufacturing processes were performed by Kuwotech Corporation (Gwangju, Korea), and the preforms were donated to the Department of Periodontology, Chonnam National University, College of Dentistry for the experiment. Disks were cleaned with 70% ethanol for 10 minutes and washed with distilled water for 20 minutes in an ultrasonic bath (Sonorex, Bandelin, Germany). This cleaning cycle was repeated three times. The disks were then rinsed with distilled water and dried under laminar airflow (Fig. 1).

Fig. 1. Tricalcium phosphate (TCP), hydroxyapatite (HA), biphasic calcium phosphate (BCP) disk, and BCP screw. (A) TCP, HA, and BCP disk (left to right), (B) BCP screw.

Surface characteristics

Observation of the surface morphology and surface composition

Surface morphology and composition were analyzed via scanning electron microscopy (SEM, S-4700; Hitachi, Tokyo, Japan) and energy dispersive X-ray spectroscopy (EDX, Emax; Horiba, Kyoto, Japan). The surface composition was measured in three areas per specimen for each group. The average composition of three measurements was calculated. An ion sputter coater (E-1030, Hitachi; Tokyo, Japan) sputtered gold–palladium (Au/Pd) on all polished disks. Subsequently, the disks were observed via SEM. SEM images of all groups were acquired three times at ×1,000, ×5,000, and ×10,000 magnifications.

Evaluation of roughness

An electronic portable surface roughness (Ra) tester (Diavite DH-8; ASMETO AG, B^ülach, Switzerland) was used to measure the Ra. Ra was calculated using the mean value of the perpendicular measurements from different areas on the surface of each disk.

Evaluation of chemical composition and crystal form of BCP

X-ray diffractometry (XRD, D/MAX Uitima III; Rigaku, Osaka, Japan) was performed to evaluate the chemical composition and crystal form. A CuKα incident radiation with a current of 40 mA and tube voltage of 40 kV was used. The scanning speed was 2°/min, and the scanning angle ranged from 20° to 90°.

Biological characteristics

Cell culture

MC3T3-E1 cells (ATCC, Rockville, MD, USA), derived from mouse calvarium tissue, were cultured in alpha minimum essential medium (α-MEM; Invitrogen Co., Waltham, MA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen Co., Waltham, MA, USA), 100 µg/mL penicillin, and 100 µg/mL streptomycin at 37℃ in a humidified atmosphere of 5% CO₂.

Observation of cell attachment and spreading using SEM

To evaluate the morphology of attachment and early growth of MC3T3-E1 cells, disks seeded with MC3T3-E1 cells were observed using SEM image. MC3T3-E1 cells were seeded at a density of 1×10⁴ cells/mL with α-MEM media containing 10% FBS. The cells were incubated for 2 days, rinsed with phosphate-buffered saline, fixed with 2.5% glutaraldehyde in 100 mM cacodylate buffer (Sigma-Aldrich Korea, Seoul, Korea), and dehydrated in increasing concentrations of ethanol (30%, 60%, 95%, and 100%). Disks were immersed in hexamethyldisilazane (Sigma-Aldrich Korea, Seoul, Korea) for 15 minutes, mounted on aluminum stubs, and immediately coated with Au/Pd alloy. SEM image was acquired twice in each group at ×250 and ×500 magnifications.

Cell viability test

3-(4,5-Dimethylthiazol–2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay was performed on day 1, 3, and 5 after cell seeding to evaluate early cell viability. MC3T3-E1 cells were seeded on TCP, HA, and BCP disks placed in a 24-well plate at a density of 2×10⁴ cells/mL in α-MEM solution. MTT assay (CellTiter 96 AQueous Solution; Promega, Madison, WI, USA) was performed on day 1 and 3 of incubation. Cell viability was analyzed by determining formazan accumulation by measuring absorbance at 570 nm using a microplate reader (VERSAmax; Dynamic Device, Wilmington, DL, USA). All MTT assays were performed in triplicate.

Animal study

Beagle dog model

Two male beagles, aged 18 months and weighing approximately 10 kg (Central Lab Animal Inc., Seoul, Korea), were used in this study. The experimental protocol was approved by the Committee on Animal Research and Ethics of Chonnam National University (CNU IACUC-YB-2017-26).

Surgical procedure and sacrifice of beagles

BCP screws were inserted into the body of the mandible using a special tool designed by Kuwotech Corporation. A screw driver and bur for a 1mm diameter BCP screw were manufactured. The surgical procedures were performed under general and local anesthesia. Following sulcular incisions and reflection of mucoperiosteal flap, mandibular third premolars were extracted. A hole was formed on the body of mandible with a bur specially designed for BCP screw insertion with a diameter of 1 mm and a length of 4 mm, and the BCP screw was inserted with a screw driver (Fig. 2). Then, it was sutured with 4-0 vicryl silk. All TCP and HA screws were fractured upon insertion after drilling a hole in the alveolar bone; thus, comparative experiments with TCP and HA screws could not be conducted. After the surgical procedure, antibiotics (cefazoline sodium, 20 mg/kg) and analgesics (caprofen, 4.4 mg/kg) were administered for postsurgical infection control for the first 1 week following surgery. Sutures were removed after 2 weeks. Plaque control was maintained daily by topical application of the 2% chlorhexidine solution. The dogs were euthanized with an overdose of sodium pentobarbital (90 mg/kg) 8 weeks after the surgical procedure.

Fig. 2. Placement of biphasic calcium phosphate screws in the mandible.

Histological analysis

Tissue blocks including screws, bones, and soft tissues were resected, rinsed in saline and fixed in 10% buffered formalin. Blocks were dehydrated through a series of ethanol solutions of increasing concentrations, infiltrated in resin, and embedded and polymerized in resin blocks. Implanted screws were cut perpendicular to the axis of the screw and reduced to a thickness of approximately 30 µm using a grinding machine (EXAKT cutting/grinding systems; EXAKT Advanced Technologies GmbH, Norderstedt, Germany). The sections were stained using Goldner’s trichrome and observed under a light microscope (BX50; Olympus Optical, Osaka, Japan).

Statistical analysis

One-way analysis of variance followed by Scheffe’s test was used to assess the surface Ra and cell proliferation. SPSS software (SPSS statistics, version 12.0; IBM Co., Chicago, IL, USA) was used to determine the significance of differences, and p-values <0.05 were considered significant.

Surface characteristics

Observation of the surface morphology and surface composition

The surface morphologies of all groups are shown in Fig. 3. TCP-HA composite (BCP) showed a moderate pattern of TCP and HA. In SEM, TCP showed a slightly sparse appearance at low magnification, and particles were observed to be connected at high magnification. A microsphere was also observed between particles at high magnification. HA showed a more compact array than TCP at low magnification. The size and number of porosity were smaller than that of TCP at high magnification. BCP showed a pattern similar to HA.

Fig. 3. Surface morphology of tricalcium phosphate (TCP), hydroxyapatite (HA) and biphasic calcium phosphate (BCP) by SEM (A: ×1,000; B: ×5,000; C: ×10,000).

Fig. 4 shows the composition of TCP, HA, and BCP. The calcium and phosphorus (Ca/P) ratios of TCP, HA, and BCP were 1.58, 1.80, and 1.77 (by weight), respectively. Considering that the Ca/P ratio in the bone is 1.67, the Ca/P ratio of TCP in this study was slightly lower and that of BCP was slightly higher than natural bone.

Fig. 4. Surface composition by energy dispersive X-ray spectroscopy of tricalcium phosphate (TCP), hydroxyapatite (HA), and biphasic calcium phosphate (BCP). (A) Composition by EDX of TCP. (B) Composition by EDX of HA. (C) Composition by EDX of BCP.

Roughness test

The results of the roughness test in all groups are shown in Table 1. The Ra values of TCP, HA, and BCP were 1.63, 0.38, and 0.40 µm, respectively. TCP had the roughest surface, and HA and BCP showed similar roughness. Ra demonstrated significant differences, and post hoc analysis revealed significant differences among all groups, except HA and BCP.

XRD analysis and crystal form of BCP

The XRD patterns of BCP are shown in Fig. 5. XRD peak analysis confirmed that BCP was mixed with HA and TCP. The XRD analysis revealed a successful mixture of components.

Fig. 5. X-ray diffractometry (XRD) patterns of biphasic calcium phosphate (BCP). The XRD patterns of BCP showed that BCP was successfully mixed with hydroxyapatite (HA) and tricalcium phosphate (TCP).

Fig. 6 shows the crystal form of BCP after heat treatment. Fig. 6A presents the SEM image before heat treatment of BCP, and Fig. 6B is the SEM image after heat treatment at 850℃. BCP screws were sintered after heat treatment, and sintering resulted in increased grain coarsening and densification.

Fig. 6. Crystal form of biphasic calcium phosphate (BCP) after heat treatment. (A) SEM image before the heat treatment of BCP. (B) SEM image after heat treatment at 850°C (×30,000).

SEM analysis of osteoblastic cell proliferation

SEM images of the attachment of MC3T3-E1 cells and immortalized cell line such as pre-osteoblast to the surface of all disks after 2 days of culture are shown in Fig. 7. MC3T3-E1 cells were scattered on and attached to the TCP surface. At high magnification, the cells had an oval and spindle shape and were connected by projections as shown in Fig. 7A, A’. More cells were attached to HA and BCP surfaces than to the TCP surface. At high magnification, the cells were flattened and elongated in a spindle shape without any shape abnormality, and they were connected to the adjacent cells by pseudopods. HA and BCP exhibited a layer of dendritic-shaped cells, with cytoplasmic extensions connected with one another, as shown in Fig. 7B, B’, C, C’.

Fig. 7. SEM image of MC3T3-E1 cell attachment (2 day observation) hydroxyapatite (HA) and biphasic calcium phosphate (BCP) group exhibited a layer of cells, dendritic shape, with cytoplasmic extensions connected with one another. (A) TCP group, (B) HA group, (C) BCP group (A, A’, B’, C’: ×250; B, C: ×500).

MTT assay

Fig. 8 shows the results of the MTT assay that was performed on day 1 and 3 after seeding. On day 1, HA and BCP demonstrated significantly higher activities than TCP; however, the activities of HA and BCP were not significantly different. On day 3, BCP showed significantly higher activity than TCP, but no significant difference was observed between HA and BCP.

Fig. 8. MTT assay of tricalcium phosphate (TCP), hydroxyapatite (HA), and biphasic calcium phosphate (BCP). *Statistical different to TCP (p<0.05).

Histological analysis

Histological analysis showed that the shape of the screw remained intact even after 8 weeks of implantation. BCP screws were well-implanted in the dog mandible. The screw was in direct contact with the bone, and no fibrous tissues were observed between the screw and bone. In addition, a new bone was formed at the head of the screw (Fig. 9).

Fig. 9. Goldner’s trichrom stain of biphasic calcium phosphate in dog mandible (A: ×5; A’, B, D: ×10; B’, C, D’: ×50; C’: ×100).

Recently, there is an increasing demand for absorbable biomaterials not only in the field of bone regeneration but also in surgery, such as fracture reduction. Although the composition of HA and b-TCP in BCP is varied, many studies have reported excellent results at a 7:3 ratio [18,19]. When the bone is fractured, reconstruction with titanium plate and screw is performed the most, providing excellent results [1-5]. However, the development of absorbent materials is necessary because metal materials require a second operation for their removal after bone healing. Furthermore, the long-term presence of metal plates and screws in the body may inevitably cause metal ion elution due to the corrosion of the metal [6-8].

In this study, a composite made of TCP and HA in a ratio of 7:3 was used to make disks and screws. Physical and biocompatibility tests were performed to examine if this absorbable screw can be used as a surgical material for fractures. To evaluate the physical properties of BCP, SEM observation, roughness test, EDX analysis and XRD analysis were performed using a 15-mm-diameter BCP disk. To evaluate biocompatibility, cell attachment pattern analysis through SEM and MTT assay analyses, and in vivo histological analysis were performed. The analysis of physical properties showed that BCP has a denser structure than TCP. The Ca/P ratio of BCP was slightly higher than that of bone, and had a smoother surface than TCP.

We also designed TCP, HA, and BCP in disk forms and used them in experiments to elucidate the cellular response of MC3T3-E1 cells to examine the biocompatibility. HA and BCP exhibited a layer of dendritic-shaped cells with cytoplasmic extensions connected with one another. MC3T3-E1 cell responses such as cell attachment, spreading, and migration are dependent on the material’s surface properties [13]. Early cell attachments and subsequent cell responses to BCP are critical and prerequisite parameters for osteogenesis [14]. For each sample, cellular morphology was examined via SEM. Cells on the BCP surface had several extensive networks of cytoplasmic processes, connection to each other through microvilli between neighboring cells, and highly flattened morphology at 48 hr. However, no significant morphological difference was observed among the groups. The absence of significant morphological modification indicated that BCP surfaces are cytocompatible.

The MTT assay is a method of determining the number of viable cells in proliferation [34]. In the MTT assay, BCP showed significantly higher activity than TCP on day 1 and 3. Two reasons were identified for why BCP showed better cell response than TCP. First, the variance in the roughness of material surfaces results in different cell responses. Cells generally tend to attach and proliferate better on smoother surfaces than on rougher ones [35]. Second, there is a density difference in the material surface. In other words, if the density of the surface is higher, the cell proliferation is more active. As the surface of TCP has a lower density and porosity, the proliferation of TCP was lower than that of BCP [36]. These results proposed that BCP has excellent biocompatibility and also with osteoblasts.

Although the biological activity and biodegradability of BCP can be controlled by adjusting the ratio of HA/β-TCP, the ratio of HA/β-TCP for optimal osteoconductivity in vivo has not been clearly suggested. In various bone defect models, there are reports that bone formation was increased [37], but there are also reports that bone formation was not as expected [38,39]. So it is difficult to get predictable results when using BCP alone. Recently, studies on applying BMP or basic fibroblast growth factor to increase the bone formation ability of BCP graft materials have been conducted and favorable results have been reported [40,41].

Although our in vitro results showed that BCP was biocompatible, further in vivo experiments are necessary to prove it. Therefore, we designed screw-type BCP and special bur for drilling hole. The screw hole is usually designed a little smaller than the screw. The screw is fixed using the elasticity of the alveolar bone during screw insertion. When a HA screw was inserted into the screw hole, the screw was fractured because it was too brittle. The TCP screw was too soft to be inserted into the hole. As a result, comparative experiments with TCP and HA screws were not possible and only BCP screws were used. The compressive strength of normal human cortical bones was reported to be between 100 and 200 MPa, and the compressive strength of cancellous bones was between 1 and 12 MPa [42]. The compressive strength of sintered HA at 900℃ was approximately 285 MPa [43]. When sintering β-TCP alone, it is difficult to strengthen its mechanical properties because it is converted to α-TCP at a high temperature [44]. In this study, the mean compressive strengths of the 4- and 6-mm-diameter BCP screws were 57.2 (±8.8) and 71.6 (±23.4) MPa, respectively (data unshown). Considering the previous report, the compressive strength of BCP screws was lower than that of HA screws, but it was sufficient for clinical use. Histological analysis at 8 weeks after the insertion of BCP screws into the mandible of beagles showed osseointegration between the screw and bone. The shape of the BCP screws was maintained well after 8 weeks of implant placement. The shape remained intact for 8 weeks, which usually requires 12 weeks for bone union [45].

In this study, BCP screws were inserted directly into the mandible of a large animal to achieve the reaction between the screw and the compact and cancellous bones. In previous studies, absorbable screws were used in surgery, but reported that the rate of absorption is so fast that it could be a failure factor for bone reconstruction [22,23]. However, in our study, BCP screws were retained without disintegration until 8 weeks, suggesting that BCP screws can be used as absorbable screws in surgery. Bone formation was observed on upper side of the inserted BCP screw in histological analysis. These results propose that BCP screw has an osteoconductive property. This property might be an advantage when BCP is used for tenting screws in GBR.

As a result of this study, it was found that the screw-type BCP was well integrated into the bone. Thus, this study indicates the possibility of the clinical use of BCP screws. However, this is the first step for clinical application. Before clinical trials begin, furthder additional experiments such as BCP toxicity studies in accredited institutions and animal studies for long term observation are necessary to determine hypersensitivity, the time required for BCP screw absorption, and any side effects.

Conventionally, for the treatment of fracture, metal screws and plates were mainly used to fix the bone after the removal of bone-related lesions. They were also used to maintain space for bone regeneration; however, they have disadvantages such as discomfort of secondary operation and dissolution of metal ions in vivo. Therefore, it is necessary to develop absorbable screws and plates.

TCP and HA are known to be highly biostable as synthetic bone materials. TCP-HA composite (BCP) is mainly used in clinical practice as bone graft material. In recent years, in vivo experiments have been conducted by using BCP to make screws. In this study, BCP was prepared in the ratio of 7:3, and a 15-mm BCP disk was prepared for in vitro experiments. After that, BCP was used to make a screw to be implanted into the mandible of beagles.

The results showed that BCP screws were successfully osseointegrated to the mandible of beagles and suggested excellent mechanical properties and biocompatibility of BCP. Depending on its processing method, BCP could be used as screws and plates for fracture reduction or as membranes and tents for GBR.

None.

The authors declare that they have no competing interests.