1. Introduction
Edentulism is a disabling condition that significantly affects patients’ functional and psychosocial well-being [
1]. Although conventional complete dentures have been the primary treatment for more than a century, they are increasingly being replaced by implant-supported removable or fixed complete dentures, which provide superior comfort, function, and patient satisfaction [
2,
3].
Progressive resorption of the alveolar ridge can impose structural limitations that complicate implant placement. When bone volume is inadequate for implant-supported rehabilitation in the posterior regions, instead of advanced surgical techniques, such as various bone augmentation techniques, the use of short implants, or long cantilever restorations, implants can be tilted passing the maxillary sinus in maxilla or mental foramen in the mandible. In the completely edentulous jaw, positioning of four implants by tilting the most posterior ones between the mental foramina or maxillary sinus is a well-documented protocol demonstrating high clinical success rates [
4,
5]. Tilting the posterior implants reduces cantilever length and increases the bone–implant contact area, thereby improving stress distribution and minimizing peri-implant bone loading [
6]. The concept of tilting posterior dental implants by up to 45° relative to the occlusal plane was first introduced by Malo and is known as the All-on-4 concept [
7].
A rigid framework that splints these implants further enhances biomechanical stability by evenly transferring functional stresses [
6]. Although angled multi-unit abutments are commonly used for the tilted distal implants in this concept, micro-gaps at the abutment–implant interface can harbor bacteria and toxins, predisposing patients to peri-implant inflammation [
8,
9]. Additionally, abutment screw loosening and fractures are frequent mechanical complications [
10]. To eliminate such issues, bendable monoblock implants that integrate the abutment and the implant body into a single unit have been introduced [
11]. Bendable monoblock implants offer significant advantages in All-on-4 prosthetic applications due to their inherent flexibility, which allows for better adaptation to the patient’s anatomical variations and reduces the risk of stress concentration at the implant interface [
11].
The long-term success of implant-supported fixed complete dentures depends on understanding the underlying biomechanical principles of implantology to avoid bone overload and implant failure [
12,
13,
14]. The geometry and material of the prosthetic framework play a crucial role in the distribution of mechanical stresses to the peri-implant bone [
15]. The framework unites the implants into a single structure, enabling uniform transmission of functional loads [
16].
Due to its low density, favorable mechanical properties, and biocompatibility, titanium (Ti) alloy is widely used to fabricate the framework [
17]. However, Ti alloy frameworks have several drawbacks, including aesthetic limitations, casting difficulties, metallic taste, and incompatibility with imaging modalities [
18]. Additionally, the high elastic modulus of Ti alloy (110 GPa) compared to that of bone (14 GPa) may induce stress shielding, which can potentially contribute to marginal bone loss and implant failure [
18,
19].
To address these concerns, polymeric frameworks, manufactured using computer-aided design/computer-aided manufacturing (CAD/CAM) technology, have been proposed as alternatives to frameworks fabricated using materials such as Ti alloy, cobalt-chromium (Co-Cr), and zirconia (ZrO
2) [
16,
20,
21,
22]. High-performance polymers, such as polyetheretherketone (PEEK) and polyetherketoneketone (PEKK), offer several advantages, including low weight, low cost, and high shock absorption capacity. Their elastic modulus (~3.5 GPa for PEEK and 5.1 GPa for PEKK) is comparable to that of the human maxillary bone (cortical bone: ~13.7 GPa; trabecular bone: ~1.37 GPa) [
23,
24]. PEKK has a higher proportion of ketone groups, which provides superior thermal stability, greater potential for surface modification, and enhanced mechanical performance compared to PEEK [
25]. Another promising alternative material is a glass fiber-reinforced polymer composite (FRC), which has an elastic modulus (~35 GPa) [
26] lower than that of Co–Cr alloy (~218 GPa) [
27] but higher than that of PEEK (3.5 GPa) [
23].
Materials with high elastic moduli, such as metallic frameworks, transmit larger stresses to the bone-implant interface because they do not absorb occlusal shock [
28,
29]. In contrast, materials with lower moduli can act as stress breakers by dissipating occlusal forces and reducing stress concentration at the implant-bone interface [
22]. Nonetheless, several studies suggest that a rigid framework may provide more favorable load distribution and reduce the risk of peri-implant overload [
24,
30,
31].
The finite element analysis (FEA) is widely used in implant biomechanics due to its ability to simulate complex clinical conditions with high precision [
32]. Given the increasing interest in high-performance polymers as alternatives to metals for fabricating the frameworks, evaluating their biomechanical behavior in implant-supported fixed complete dentures has become essential. However, the literature presents inconsistent findings regarding the effect of framework material rigidity on stress distribution in implant-supported prostheses. While some studies [
33,
34] have shown that polymer-based materials with lower elastic modulus increase stress transfer to the implant and the surrounding bone, others [
35,
36] have reported reduced stress levels in the peri-implant bone associated with the use of less rigid or semi-rigid materials, which may provide more uniform load distribution.
In addition, limited information is available regarding the biomechanical performance of different framework materials in implant-supported fixed complete dentures supported by bendable monoblock implants following the All-on-4 concept. Because monoblock implants eliminate the conventional implant–abutment interface, the mechanical behavior of implant-supported fixed complete dentures supported by such implants may differ from that of conventional restorations. Therefore, evaluating the effect of framework material selection in such configurations is clinically relevant. The purpose of the present study was to use FEA to determine the influence of framework material on stresses in the cortical bone, the bendable monoblock implant, screws, and the framework in a three-dimensional (3D) model that comprised a fixed complete denture supported by four implants arranged in a configuration that follows the All-on-4 concept. The framework materials used were Ti alloy, Co-Cr, PEEK, PEKK, and FRC. The null hypothesis was that the framework material would not affect the stress distribution patterns in either the bone or the bendable implants.
2. Materials and Methods
The study protocol was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Istanbul Yeni Yuzyil University Science and Health Sciences Research Ethics Committee (Approval No. 2023/10-1100, Approval Date: 2 October 2023).
The generation of the 3D mesh, its transformation into a mathematically optimized solid model, the construction of the finite element models, and the FEA were all performed on an HP workstation (Intel Xeon E-2286, 2.40 GHz, 64 GB ECC memory). Bone structure was modeled using tomographic data obtained from the Visible Human Project (The National Library of Medicine, FACT SHEETS Office of Communications and Public Liaison National Library of Medicine, Bethesda, MD, USA), a publicly accessible and fully anonymized anatomical dataset, and, thus, requiring individual informed consent was not required. Bone structures were segmented from computed tomography (CT) data using 3DSlicer software (version 5.10.0). STL models were extracted from CT data via 3DSlicer (
Figure 1). Reverse engineering and 3D CAD modeling were performed using ANSYS SpaceClaim (Version 22.0), while model preparation and optimized meshing were performed in ANSYS Workbench. The finite element simulations were solved using the LS-DYNA solver.
2.1. Modeling of Cortical and Trabecular Bone
The mandibular bone model used in the study was obtained from the CT scan of an edentulous adult. Dataset was obtained from Visible Human Project. The subject was female; however, age information was not available. The CT scan was reconstructed with a slice thickness of 0.1 mm, and the resulting data were imported in DICOM (.dcm) format into 3DSlicer software. Based on appropriate Hounsfield Unit (HU) thresholds, the DICOM data were segmented to produce three-dimensional representations of the bone structures (Threshold range: 426.50–3193.04). Unnecessary regions and artifacts were removed using the “Erase” and “Scissors” tools. These models were then exported in STL (.stl) format. These 3D models were imported into ANSYS SpaceClaim, and a 2 mm cortical bone layer was created by applying a 2 mm offset to the mandibular model. The internal surface of the cortical shell was used as a reference to generate the trabecular bone model (
Figure 2). All models were positioned in the 3D coordinate space using ANSYS SpaceClaim to ensure anatomical alignment, completing the modeling phase.
2.2. Model Construction
Three-dimensional (3D) CAD models of bendable monoblock implants (Mode Provo-S, Mode Implant/Mode Medikal San. Tic. Ltd. Şti., İstanbul, Türkiye), were obtained from the manufacturer (Mode Implant/Mode Medikal San. Tic. Ltd. Şti., İstanbul, Türkiye). Two anterior implants (3.5 mm in diameter and 12 mm in length) were placed vertically in the lateral incisor regions, while two posterior implants (4.0 mm in diameter and 15 mm in length) were positioned at a 30° relative to the occlusal plane [
37] in the second premolar regions (
Figure 3).
Occlusal screws were modeled in ANSYS SpaceClaim software based on the dimensions specified in the manufacturer’s catalog. The framework and prosthetic structures were also designed using ANSYS SpaceClaim. To ensure proper load transfer between the components, mesh compatibility adjustments were made in ANSYS Workbench (
Figure 4).
Implant-supported screw-retained fixed complete denture with frameworks made of Co-Cr, Ti alloy, PEEK, PEKK, or FRC were connected to implants via occlusal screws, and separate models were constructed for each framework material. Each prosthetic superstructure consisted of 12 teeth. In the polymer-based group (PEEK, PEKK, and FRC), the superstructure was veneered with light-cure composite resin, while in the metal groups (Co-Cr and Ti alloy), porcelain was used as the veneering material (
Figure 5 and
Table 1).
The bar structure was designed with a width of 5 mm, a height of 5 mm, and a cantilever length of 10 mm [
35,
37,
38]. In the final prosthetic configuration, the cantilever was extended to 15 mm [
39], and the vertical distance between the prosthesis and the mucosal surface was also set at 15 mm [
35].
2.3. Generation of Mathematical Models
Mathematical models were generated by discretizing the geometric structures into small, simple elements in a process known as meshing (
Figure 6). The three-dimensional geometrical models developed in ANSYS SpaceClaim were imported into ANSYS Workbench and converted into numerical models suitable for FEA. In the creation of the mathematical models (mesh structures), tria (triangular) mesh sizes ranging from 0.05 to 0.25 mm (highly refined) were used. After the surface meshes of all models were generated using tria, the solid meshes of the objects were created using tetrahedral (regular four-sided) solid mesh elements. The posterior and inferior regions of the bone were fixed in all three spatial directions. All components were defined using bonded contact conditions, thereby assuming perfect osseointegration at the bone–implant interface. Subsequently, these models were exported to the LS-DYNA solver for computational analysis. Linear elastic material behavior was assumed for all analyses; as such, each material was defined by its elastic modulus and Poisson’s ratio. The mechanical properties of all materials were obtained from previously published studies [
23,
24,
26,
27,
37,
40,
41,
42] (
Table 2).
2.4. Mesh Convergence Test
A mesh convergence analysis was performed to ensure the reliability and accuracy of the finite element model used in the biomechanical analysis. The primary objective was to determine an appropriate mesh density that would achieve a relative error below 3% while maintaining a balance between computational efficiency and solution accuracy.
Accordingly, a series of finite element meshes with varying element sizes, ranging from coarse to fine, were generated. To ensure consistency, each mesh was analyzed under identical loading (
Table 3) and boundary conditions, which are described in detail in the following section. For the mesh convergence analysis, the maximum von Mises stress in the implant was used as the evaluation parameter. The results obtained from successive mesh refinements were compared, and variations in the evaluation metric were observed. The relative error in the stress obtained between use of consecutive meshes was calculated using the following equation:
During model preparation, triangular 2D and tetrahedral 3D meshes were used. These mesh types are considered suitable for modeling complex geometries and curved surfaces, such as bone structures. Mesh quality was evaluated for all models, and elements with a skewness value greater than 80° or a minimum element size below 0.001 were reviewed. Meshes did not conform to these criteria were appropriately corrected and included in the analysis.
2.5. Loading and Boundary Conditions
In fixed prostheses following the “All-on-4” concept, the average occlusal force in the premolar and molar regions has been reported to be approximately 200–300 N [
43]. In the present study, to simulate different masticatory conditions, loads were applied to all models under unilateral oblique, unilateral vertical, bilateral oblique, and bilateral vertical loads scenarios. In the oblique loading condition, a 300 N force was applied at a 45° angle from the lingual to the buccal direction on the buccal cusps of the first premolar, second premolar, and first molar teeth. In the vertical loading condition, a 300 N force was applied perpendicularly to the same regions (
Figure 7). The occlusal load was distributed on the three posterior teeth.
The applied forces were distributed over adjacent nodes to avoid stress singularities at the loading regions. All models were constrained at the posterior and inferior regions of the mandible by fixing the corresponding nodes and restricting all degrees of freedom in the X, Y, and Z directions. Additionally, symmetric boundary conditions were applied to all model components in the Y–Z plane, normal to the
X-axis (
Figure 8). Under these loading and boundary conditions, a total of 20 linear static analyses were conducted across the five framework material models (
Table 4).
2.6. Stress Distribution Analysis
In this study, stress criteria were selected based on the mechanical behavior of each material. All prosthetic components (implants and screws) and framework structures in the “All-on-4” system were analyzed using von Mises equivalent stress, which is appropriate for ductile materials. For cortical bone, the maximum (tensile) (σ
max) and minimum (compressive) (σ
min) principal stresses were calculated according to the failure theory of maximum principal stress. This method was preferred because bone exhibits lower tolerance to tensile loads than to compressive loads and exhibits behavior similar to brittle materials [
23].
4. Discussion
Understanding stress distribution at the implant-bone interface is essential for ensuring the long-term stability of dental implants. The magnitude of stress transmitted from a prosthetic framework to the surrounding bone is influenced by the elastic modulus of the framework material [
44]. While rigid metallic frameworks remain the gold standard for implant-supported fixed complete dentures, there is limited information regarding the biomechanical behavior of high-performance polymer-based materials, particularly when used with bendable monoblock implants. Most current investigations are restricted to a single material group or standardized, simplified occlusal scenarios [
45,
46]. This study aimed to clarify the biomechanical performance of FRC, under unilateral oblique, unilateral vertical, bilateral oblique, and bilateral vertical loading conditions, in comparison with PEEK, PEKK, and conventional metallic alloys. The results indicated that polymer-based frameworks with lower elastic moduli transmitted noticeably higher stresses to the cortical bone and implants. As such, the null hypothesis of the study was rejected.
The results demonstrated that framework material and loading configuration significantly influenced stress distribution across cortical bone, implants, screws, and framework structures. PEEK and PEKK frameworks exhibited the lowest stress levels within the framework, indicating effective stress-absorbing behavior; however, this was accompanied by increased stress concentrations in cortical bone, implants, and screws, reflecting load transfer to the supporting components. This behavior is attributed to the low elastic modulus of these materials, which allows greater framework deflection and shifts stress toward the implant–abutment interface. While flexible frameworks reduce internal stress, they may increase stress at critical interfaces, potentially predisposing to screw loosening or abutment fatigue [
47,
48]. Conversely, Co–Cr and Ti alloy frameworks exhibited the highest stress values within the framework due to their high rigidity, while transmitting relatively lower stress to cortical bone, implants, and screws. FRC demonstrated an intermediate biomechanical response, providing a balanced stress distribution between structural rigidity and load transfer to surrounding tissues. This intermediate behavior can be explained by the materials elastic modulus (approximately 35 GPa), which lies between low-modulus polymers (PEEK: ~3.5 GPa; PEKK: ~5.1 GPa) and high-modulus alloys (Ti alloy: ~110 GPa; Co–Cr: ~218 GPa), which allows partial stress absorption while limiting excessive stress transfer to surrounding structures. In addition, the loading configuration played a critical role in determining the magnitude of stress. Oblique loading consistently produced higher stress values than vertical loading across all components, likely due to the presence of lateral force and bending moment. Furthermore, posterior implant regions exhibited more pronounced stress variations, particularly under unilateral loading conditions, which may be associated with cantilever effects and asymmetric load distribution.
Some of the trends in the results in the present work are the same as those reported in the relevant literature studies. Dayan and Geckili [
47] reported that PEEK and PEKK frameworks exhibited lower internal stresses while transferring greater stress to surrounding bone and implant components. Güzelce [
26] reported that Co-Cr frameworks displayed higher internal stress, resulting in lower stress values in the supporting bone and implants. Villefort et al. [
49] compared PEKK and PEEK and found that PEKK frameworks exhibited lower stress concentrations at the prosthetic screw than PEEK, suggesting a reduced risk of screw loosening. Meghna et al. [
48] reported that the highest stress values in the framework were observed in the Co-Cr group, whereas the lowest stress values were observed in the PEEK group. However, unlike the findings of the present study, they also reported that the lowest stress values in the bone were observed in the PEEK group. de Araújo Nobre et al. [
50] reported that the use of PEEK frameworks and hybrid PEEK-acrylic resin prostheses in All-on-4 restorations reduces marginal bone loss and improved long-term biological outcomes, which was attributed to the shock-absorbing properties of PEEK. Similarly, Chen et al. [
40] found that the use of a PEEK framework provided more favorable stress distribution in the jawbone due to its lower elastic modulus. Cabbarova et al. [
46] reported that materials with low elastic modulus (PEEK and PEKK) increased stress transmission to peri-implant bone and connection components, whereas rigid materials such as Ti alloy and zirconia provided a more balanced load distribution with lower stress concentrations. In addition, FRC frameworks remained within clinically acceptable biomechanical limits. Similarly, Lahoud et al. [
51] reported that FRC exhibited higher fracture resistance than zirconia, indicating its potential as a reliable alternative for full-arch rehabilitations. Demirci et al. [
52] also found that FRC demonstrated significantly greater fracture loads compared to PEEK and PEKK, while Franco et al. [
53] showed that FRC frameworks achieved a more uniform occlusal load distribution around implants compared to rigid metallic structures. In line with these findings, the present study demonstrated that the FRC framework exhibited lower σ
max and σ
min stress values in cortical bone compared to polymer-based framework and higher values compared to metal-based framework under bilateral and unilateral loading conditions, revealing that the FRC framework demonstrated a balanced biomechanical response. In the present study, conventional multi-unit abutments were replaced with bendable monoblock implants specifically designed for compatibility with the All-on-4 configuration. Monoblock implants eliminate both the abutment screw and the microgap at the implant-abutment interface, which can reduce stress concentrations in the connection region. Zincir and Parlar [
3] reported that monoblock implants resulted in lower stress levels around the implant and peri-implant bone compared to traditional systems. The results of the current study further support these observations, indicating that monoblock configurations promote a more uniform stress distribution, potentially minimizing marginal bone loss and enhancing long-term biomechanical stability. The maximum von Mises stress values in metallic frameworks (Co-Cr and Ti alloy) were primarily concentrated within the framework structure but remained well below the yield strengths of these materials [
54,
55]. In our study, the maximum tensile stress (σ
max) values in cortical bone remained below the allowable limit of 66 MPa across all groups and loading conditions [
35]. Notably, under certain loading scenarios, σ
max values in the anterior region surpassed those in the posterior, which has clinical relevance given the limited tensile strength of cortical bone. However, since all values remained within safe limits, we conclude that no structural risks were present. When examining minimum compressive stress (σ
min) values, relatively low stress levels were observed in the anterior regions, indicating that loads were primarily transferred to the posterior implants.
The findings of this study contribute to the evolving understanding of stress distribution at the implant-bone interface, particularly regarding the comparative performance of PEEK, PEKK, and FRC frameworks versus metallic frameworks. The observed differences in stress transfer mechanisms underscore the importance of material selection in optimizing implant design and enhancing the longevity of dental restorations. Further research is essential to explore the long-term implications of these materials in clinical practice. To enhance the clinical relevance of future investigations, additional in vitro and in vivo studies are necessary. These studies should examine the hybrid biomechanical performance of FRC frameworks under dynamic loading conditions while considering varying implant angulations, interface configurations, and load application types.
This study has several limitations. The finite element model assumed that all materials were isotropic, homogeneous, and linearly elastic, which does not fully represent the complex biomechanical behavior of biological tissues. Additionally, the reliance on anatomical data from a single patient restricts the generalizability of the results. The study utilized static loading conditions, failing to account for the dynamic nature of actual chewing activities. The long-term performance of materials under cyclic loading and potential fatigue failure were not assessed. The study overlooked biological factors that could influence implant success, such as individual healing responses and bone remodeling processes; in addition, different materials were used as the veneer agent on the superstructure (the 12 teeth), which may influence stress distribution patterns. Moreover, different framework materials not included in this study may affect biomechanical behavior, and further studies evaluating different material combinations are needed. Another limitation is that important biomechanical responses, such as displacement and micromotion of implants and prosthetic components, were not analyzed. Lastly, the results of this study are specific to a fixed set of parameters, including implant design, posterior implant angulation, and cantilever length, and variations in these factors may significantly influence stress distribution patterns.