1. Introduction
Several factors influencing the osseointegration of dental implants include biocompatibility of implant material, implant surface and design, healthy implant bed and good bone quality, surgical technique, adequate healing phase, and loading conditions [
1]. Among these, the implant material, surface topography, and design are factors that may be controlled during the manufacturing process of the dental implant. The vast majority of currently available dental implants are made of commercially pure (grade 4) titanium. Because of increased failure with pure titanium implants in areas subjected to high wear or tensile strength, titanium alloy implants have emerged as an alternative choice, owing to their enhanced mechanical characteristics [
2]. The reason for titanium and its alloys being the materials of choice for dental implants is their excellent biocompatibility. Among the titanium alloys for biomedical applications, the system titanium-aluminum-vanadium (Ti
6Al
4V; grade V titanium) is commonly used in orthopedic and dental implants. Adding aluminum and vanadium to titanium significantly increases the strength, ductility, and fatigue resistance of the material [
2].
The elastic modulus is an important measure for biomechanical interaction between the implant and the bone. Although Ti
6Al
4V exhibits superior mechanical properties and comparable biocompatibility to conventional pure titanium, it presents a slightly higher elastic modulus. In the case of orthopedic prostheses, a mismatch of elastic modulus values could result in stress shielding that leads to undesirable bone resorption around the implant and subsequent aseptic loosening of the implant [
3]. The reported Young’s modulus values in the literature range between 6.9–25 GPa for bone, 103–107 GPa for commercially pure titanium, and 114–120 GPa for Ti
6Al
4V [
4]. It is very likely that the stress shielding phenomenon in conjunction with the complicated oral environment may contribute to the progressive peri-implant bone loss beyond the initial physiologic bone remodeling. An ideal dental implant is expected to possess a combination of high strength and low elastic modulus that make it strong enough to prevent deformation during insertion, and to withstand loading from tension, compression, bending, or torsion, to assure stability between the implant and prosthetic components, yet it must have low stiffness to avoid shielding the bone from stress. Unfortunately, these characteristics seem contradictory for solid metal materials. Fabrication of open-cell porous titanium alloy implants may be a good solution that reduces the effective elastic modulus to minimize the elastic mismatch, allowing for bone ingrowth inside the implant to enhance fixation [
5].
Implant design/macrogeometry and the related surgical instrumentation technique/dimension are important parameters for attaining primary stability and successful osseointegration [
6,
7]. Studies investigating the effect of dental implant macrodesign on osseointegration mainly focus on design features, including implant size, body shape, thread design, crestal module, and microthreads [
6,
7]. Several methods have also been advocated to modify implant surface roughness for enhancing osseointegration [
8]. Surface roughness may be divided into three levels, namely macro-, micro-, and nano-scale roughness [
9]. Nowadays, commercially available dental implants are mostly fabricated by machining titanium rods, followed by modification techniques that produce a micro- or nano-rough implant surface on a high-density titanium structure. It is noteworthy that the macrogeometric design of an implant also increases the micro- and nano-level contribution to osseointegration [
10]. Until recently, several methods have been developed for manufacturing porous titanium or titanium alloy implants to promote ingrowth of bone beyond the surface [
4]. Direct metal laser sintering (DMLS) is a three-dimensional (3D) printing additive technique, in which a 3D porous structure may be built layer by layer using a high-powered laser beam to fuse powdered metal according to a predefined computer-assisted-design (CAD) file [
11]. Recent studies have demonstrated the potential use of DMLS titanium alloy dental implants [
11]. However, these DMLS dental implants are mostly designed with a very superficial porous, but non-scaffold surface, and a dense titanium core. As porosity and pore size of scaffolds for bone regeneration have been recognized to directly affect bone ingrowth [
12], the objective of this study was to optimize DMLS-manufactured Ti
6Al
4V dental implant design, minimize elastic mismatch, allow maximal bone ingrowth, and improve long-term fixation of the implant.
2. Materials and Methods
2.1. Fabrication of Laser-Sintered Titanium Dental Implants
Seven types of laser-sintered dental implant specimens (4.3 mm in diameter; 10.0 mm in length) were produced using the EOSINT M 280 system (EOS GmbH, Munich, Germany). Briefly, the process computer sliced the predefined CAD file into a stack of thin slices. A uniform thin layer of Ti6Al4V powder (20–50 μm) was deposited using a moving wiper. The computer scanned the laser beam on the powder bed surface under a controlled hypoxia atmosphere (O2 < 10,000 ppm) to bond the loose powder together. The process was repeated, and by altering the shape of each scan layer, a part of arbitrary shape could be produced. All specimens were produced directly and fully automatically from 3D CAD data without any tooling.
2.2. Characterization of the Porous Biomimetic Ti6Al4V Dental Implants
The basic shapes of the dental implants were identical to the commercially available Screw-Vent® Implant System (Zimmer Dental, Carlsbad, CA, USA) but with varying surface characteristics: arrangement of pores, size of pores, and porosity. The external pore dimensions were analyzed by optical microscope. Twenty-four pores, 4 in each of the 60 degrees range of the implants’ circumference, were randomly selected and the average dimension was calculated. Scanning electron microscopy was used to study the micro structural surface characteristics and pore morphology within the structure. The porosity and strut volume of the implant were determined using a helium-purged low-pressure pycnometer (Accelerated Surface Area and Porosity System, Micromeritics® ASAP 2060 Pycnometer; Micromeritics Instruments Corporation, Norcross, GA, USA).
2.3. Evaluation of the Mechanical Properties of the Dental Implants
The stability (screw-in torque, screw-out torque, and pull-out strength) of the different DMLS dental implants was tested in synthetic cancellous bone block (Sawbones; Pacific Research Laboratories Inc., Vashon Island, WA, USA). Three-point bending tests were performed using an ElectroPulsTM Instron E3000 (Norwood, MA, USA). Failure of the DMLS dental implants was evaluated using a torsional test machine (SE MODEL 2205NS; SE Test system Co., Ltd., Taipei, Taiwan) and NTS torsional load cells (NTS TECHNOLOGY Co., Ltd., Tokyo, Japan).
2.4. In Vitro Cell Observation and Osteogenic Differentiation
2.4.1. Scanning Electron Microscopy
Ti6Al4V discs (2.0 cm diameter × 0.5 cm height) resembling the dental implant surfaces with different porosities (0, 20, 40, 60 or 80%; 300–500 μm pore size) were produced using the same DMLS technique. The human MG63 osteosarcoma cell line was provided by American Type Culture Collection (ATCC, Rockville, MD, USA; obtained from Sigma Corp.: Sigma-Aldrich; St Louis, MO, USA). Confluent MG63 (ATCC, Rockville, MD, USA) osteosarcoma cell line cultures were seeded on top of the sterilized discs. The discs with MG63 cells were placed in a culture plate for 4 h for cell adhesion and then cultured in Dulbecco’s modified Eagle’s medium (Gibco, Life Technologies Inc., Grand Island, NY, USA). Fresh medium was replaced every 2 days. The morphology of cell proliferation was observed by field emission gun SEM (FEG-SEM) (JEOL, JSM-6330F, Tokyo, Japan). At 14 days, cells were collected and subjected to subsequent tests.
2.4.2. Alkaline Phosphatase Level
Alkaline phosphatase (ALP) activity was analyzed with the SensoLyte® pNPP Alkaline Phosphatase Assay Kit (AnaSpec, Inc., Fremont, CA, USA). Briefly, ALP in biological samples was detected by enzyme-linked immunoabsorbent assay with alkaline phosphatase conjugated secondary antibody or streptavidin using colorimetric pNPP (p-Nitrophenyl phosphate phosphatase) substrate. Upon dephosphorylation, pNPP turned yellow and was detected at absorbance = 405 nm.
2.4.3. Quantitative Real-Time Polymerase Chain Reaction (PCR)
Real-time PCR was performed with TaqMan® Gene Expression Assays (Applied Biosystems, Foster City, CA, USA) according to manufacturer’s instructions, to determine the levels of the ALP and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA transcripts. The expression levels of the ALP were calculated by normalizing the mRNA level against that of GAPDH as an internal control.
2.5. In Vivo Biocompatibility Study
2.5.1. Animal Experiments
The experimental protocols were approved by the Medical College’s Animal Research Committee of the National Taiwan University. Twenty-four adult (1-year-old) female New Zealand white rabbits were randomly divided into control (non-porous) and experimental (porous) groups. The dental implants were inserted into lateral aspect of distal femurs. Both groups were randomly subdivided into two subgroups (six animals each), representing euthanasia 6 and 12 weeks after implant insertion.
2.5.2. Micro-Computed Tomography (Micro-CT) Analysis
The dental implants after 6/12 weeks of implantation was examined by micro-CT (SkyScan 1176, Bruker, Kontich, Belgium) prior to pull-out test. The images of the bone and dental implants in the specimens were acquired and reconstructed by CTvox 2.4 software (Bruker Skyscan, Konitch, Belgium).
2.5.3. Ex Vivo Mechanical Testing
An axial pull-out force was applied using a material testing machine (MTS 858, MTS Systems, Eden Prairie, MN, USA). Each implant was extracted from the distal femur at a rate of 5 mm/min until failure, which was defined as the maximum pull-out strength before the load decreased abruptly.
2.6. Statistical Analysis
Results from mechanical experiments were expressed as mean ± standard deviation and statistically analyzed by two-way analysis of variance (ANOVA). Statistical significance by Dunnett’s test was set at p < 0.05. The unpaired two-tailed student’s t-test was performed to compare differences between the groups for in vitro and ex vivo experiments. Differences were considered significant at p < 0.05. All analyses were performed using SPSS version 16.0 software (SPSS Inc., Chicago, IL, USA).
4. Discussion
Scaffold design enhances biological anchorage to the surrounding bone by allowing ingrowth of mineralized tissue into the porous network and has been a significant feature in the application of orthopaedic tissue engineering [
12]. Several methods have been proposed to produce porous metal structures [
4,
13]. More advanced techniques that have the capability of controlling the external and internal structure are additive manufacturing 3D-printing procedures including electron beam melting [
14], selective laser melting [
15], direct metal deposition [
16], and selective laser sintering [
17]. In this study, direct metal laser sintering of pre-alloyed Ti
6AlV
4 fine powder was used to additively manufacture the dental implants. The theoretical formulation assumes a block with an infinite number of pores. In our previous study on the interference screws, the surface roughness and inter-connected porous architectures enhanced better bone-tendon-implant integration and resulted in stronger biomechanical characteristics when compared to traditional screws [
18]. However, during the manufacturing process, only a finite number of pores could be built into the implants. The differences between theoretical and experimentally observed values could be attributed to the melting of titanium alloy powder at high temperatures, and the subsequent solidification by cooling caused an unevenness of the surface, leading to pore surface curvatures and corrugation (
Figure 1). While the overall porosity, the dimension of the solid structures, and pore size collectively determine the biomechanical properties of the dental implants, a linear correlation between porosity and strength does not seem to exist (
Supplementary Table S1). This may be partly due to the fact that curvatures and corrugations of pore surface result in local heterogeneities that affect peak stress [
19].
Although the exact optimal parameters have not been validated in literature, studies have revealed that 500 μm pore-sized titanium scaffolds are more beneficial for in vitro cells growth and in vivo bone formation than larger pores [
20,
21]. In general, the minimum requirement for pore size is ≈100 μm; however, pore sizes >300 μm are recommended to enhance new bone formation and the formation of capillaries [
12,
13]. A porosity of 60–70% could best fit the mechanical properties of human trabecular bone [
22]. In this study, types #3 (50–200 µm, 36% porosity), #4 (non-porous), and #6 (300–500 µm, 55% porosity) dental implants were selected for further biomechanical tests. Although pull-out strengths were not significantly different among the three designs when tested in sawbones (
Figure 2B), after 6 or 12 weeks of implantation into rabbit femurs, bone grew into the scaffold structure of the type #6 implant, and its pull-out strength increased to approximately 2.4-fold that of the non-porous implant (
Figure 5). While relatively high porosity and large pore sizes attain greater bone ingrowth, they result in diminished mechanical properties. We further augmented the weakest #6 design with a 3D-printed longitudinal strut. The longitudinal strut acted like a weight-bearing beam and improved the biomechanical characteristics of the type #6 implant without significantly compromising the porosity (
Figure 3;
Supplementary Table S3). Addition of the localized struts reduced the overall porosity from 55% to 43–46%. However, the improvement of the mechanical properties was not due to a decrease of the porosity alone, as type #6-A (averaged 46% porosity) still demonstrated overwhelmingly better performance than the type #7 dental implant (generalized 47% porosity) (
Supplementary Table S1).
Very few clinical studies on DMLS dental implants were found in current literature [
11]. These studies were conducted by the same group of researchers and the Ti
6AlV
4 dental implants were characterized by a superficial porous surface (non-scaffold) on a bulk and dense core [
23]. Despite that the original intention of fabricating these implants was to avoid stress shielding by reducing elastic modulus, the Young’s moduli of the porous surface and compact core were 77 ± 3.5 GPa and 104 ± 7.7 GPa, respectively, still much higher than that of bone. Nevertheless, these clinical studies have reported satisfactory short-term results. Two animal studies evaluated the osseointegrative properties of titanium alloy dental implants with scaffold design [
24,
25]. In a sheep model, the DMLS implants were made with a solid core and 500 μm thick macroporous shell (pore diameter approximately 500 μm) [
24]. In a rabbit model, multi-rooted selective laser melting implants were designed with 26% porosity (300 μm pores) in the cortical part and 50% porosity (400 μm pores) in the cancellous part [
25]. Despite the completely different geometric characteristics of the implants, both studies showed significant improvement of removal torque values after 8 weeks of implantation. One recent study specifically investigated the optimal pore size (comparing 0, 200, 350 and 500 μm) for scaffold Ti
6Al
4V dental implants using 2D photoelastography on bone mimicking tissue phantom and in vitro experiments [
26]. Results showed a lower stress shielding to the surrounding bone at the porous structure of the implants and that both 350 and 500 μm pore sizes presented a notable improvement in cell proliferation, attachment, and differentiation [
26]. Taken together, the parameters used in our type #6 implant (300–500 µm, 55% porosity) seem to provide an optimal potential for improving the mechanical shielding to the surrounding bones and osteoinduction of the implant itself. However, the intraoral environment is constantly changing and dental implants always experience cyclic occlusal functioning. As implants are likely to fracture upon extended use, augmentation of the porous implants with 3D-printed longitudinal weight-bearing struts could be helpful. Additional studies will be necessary to validate this proposition. The idea of incorporating scaffold design into dental implants is emerging. Dental implants with porous tantalum components have been recently developed by Zimmer Inc. [
27]. Tantalum and titanium both belong to the refractory metal group. Unlike titanium that is widely distributed and abundant on Earth, tantalum is rare and costly. In addition to the advantage of cost-saving, using DMLS technique to fabricate porous titanium alloy dental implants allows tailoring of the implants to accommodate the underlying anatomy of the individual patients.
In conclusion, this study investigated the biomechanical and biological characteristics of scaffold Ti6Al4V dental implants manufactured using DMLS addition-production technology. Our results support that implants with interconnected pores encourage bone ingrowth and mineralization to achieve better retention. Based on our results so far, parameters such as pore size 300–500 µm and approximately 55% porosity could provide an optimal potential for improving the mechanical shielding effect and osseointegration, bearing in mind that these data were tested only on a standard dental implant shape of 4.3 mm in diameter and 10.0 mm in length. As an increase in pore size or porosity reduces total strut volume, the effective stiffness and ultimate strength of the implant also decrease. To achieve the ideal dental implant requirement of a combination of high strength and low elastic modulus, we introduced a new design concept that by augmenting the implant with a longitudinal weight-bearing strut, there is a chance to reach a balance of both biologic and mechanical demands. Future studies will be necessary to evaluate the fatigue behavior of these DMLS scaffold Ti6Al4V dental implants with/without augmented weight-bearing strut(s) possibly in a simulated oral environment to understand the long-term performance of these dental implants.