Three-dimensional finite element analysis of the stability of tooth-implant supported prostheses for mandibular anterior dentition defects Authors: Jie Yu, Meiqi Wu, Maoshen Yan, Yangtian He, Xi Suo, Yiling Liu, Yufei Li, Na Li, Jing You, Ming Yan Published: 2026 ABSTRACT BackgroundWhile tooth-implant supported prostheses (TISP) are commonly used in the posterior region, few studies report on their anatomical suitability or mechanical behavior in the anterior mandible. This study analyzed CBCT data from patients with mandibular anterior edentulism and constructed three-dimensional finite element models to investigate the anatomical constraints of this region and to compare the biomechanical behavior of different restorative designs involving TISP.MethodsBased on CBCT images of 524 patients, 3D reconstruction using Mimics 21.0 was performed to measure available alveolar bone buccolingually at 1 mm, 3 mm, and 5 mm apical to the crest. Two finite element models were established to represent clinically relevant restorative strategies for extensive mandibular anterior tooth loss: an implant-implant supported fixed partial denture and a tooth-implant supported fixed partial denture. These models were compared in terms of stress distribution under different oblique loading conditions.ResultsAlveolar width generally increases with depth from the crest. The available mandibular anterior alveolar bone width is significantly influenced by gender and age. Specifically, bone width measurements were statistically significantly greater in male patients than in female patients, and the width at most measured depths showed a stable negative correlation with age. Biomechanically, stress concentrations were mainly observed at the implant-abutment neck and the crestal cortical bone, and stress increased with increasing loading angle (overjet). Within the assumptions of this simplified finite element model, the tooth-implant supported design showed a comparatively different stress distribution pattern and may provide biomechanical reference for selected mandibular anterior restorative scenarios.ConclusionsMandibular anterior tooth-implant supported and implant-implant supported restorations showed different stress distribution patterns under oblique loading. Stress was mainly concentrated at the implant-abutment neck and adjacent crestal cortical bone, and greater loading angles were associated with higher stress levels. These findings may provide comparative biomechanical information for restorative planning in selected cases, but further experimental and clinical studies are needed to confirm their clinical relevance. Background Oral implant technology has advanced significantly in recent years. As an important restorative solution for missing teeth, it offers superior comfort, aesthetics, and helps preserve natural tooth structures [1]. Among common clinical challenges in prosthodontics, mandibular anterior tooth loss holds particular importance—these teeth crucially impact both facial aesthetics and essential functions like mastication, speech articulation, and oral health maintenance [2]. Epidemiological data indicates a notably high prevalence of mandibular anterior tooth deficiency among individuals over 35 in China, with primary etiologies including periodontal disease (41%), traumatic injuries (29%), and congenital absence (18%) [3]. Conventional prosthetic approaches for anterior tooth loss have important limitations. Removable partial dentures may provide limited comfort and retention, whereas conventional fixed bridges require preparation of adjacent teeth [4]. Implant restoration is often considered a favorable option; however, in the mandibular anterior region, insufficient residual bone volume may complicate implant placement. Under such conditions, tooth-implant supported prostheses (TISP) may be considered as an alternative restorative strategy in selected cases. Nevertheless, the clinical use of TISP remains controversial because of the biomechanical mismatch between natural teeth and osseointegrated implants, and long-term outcomes reported in the literature have not been uniformly favorable [5, 6]. The mandibular anterior region presents unique anatomical and biomechanical challenges for implant restoration. After tooth loss, the residual ridge in this region often exhibits a narrow buccolingual morphology, which may limit implant placement and increase the technical sensitivity of treatment [7, 8]. In addition, although the absolute bite force in the anterior region is lower than that in the posterior region, anterior teeth are frequently subjected to oblique functional loading, and variations in overjet and occlusal relationship may substantially influence stress distribution around the implant and supporting structures [9, 10]. Current TISP research debates center on three key aspects: anatomical limitation, biomechanical mismatch, and the lack of design standardization. The first aspect is that the mandibular anterior region may present anatomical constraints that limit implant positioning and restorative flexibility [11]. Second, because natural teeth and implants differ substantially in mobility, their combined use may lead to nonuniform stress transfer and stress concentration in supporting structures [12]. Third, although canine involvement and various connection strategies have been discussed in the literature, evidence remains limited regarding the stress distribution characteristics of different restorative designs in mandibular anterior edentulous defects. Finite Element Analysis (FEA) is a computational method widely used to investigate stress and strain distributions in complex oral structures under defined loading conditions [13]. In implant dentistry, FEA has been applied to evaluate the mechanical response of teeth, alveolar bone, implants, and restorative components under simulated functional loading [14]. Meanwhile, modern CBCT imaging enables detailed characterization of alveolar morphology and residual bone availability. When combined with FEA, CBCT-based modeling can provide a useful framework for exploring the relationship between regional anatomy and restorative biomechanics [15]. Although previous studies have investigated implant biomechanics, oblique loading, and tooth-implant combined restorations, limited evidence is available for the mandibular anterior region, where narrow alveolar bone width may constrain implant placement and influence restorative design selection [7, 8]. Moreover, few studies have combined large-sample CBCT-based morphometric analysis with finite element comparison of different restorative strategies specifically for mandibular anterior edentulous area. Therefore, the aim of the present study was to: (1) characterize the buccolingual alveolar bone width of the mandibular anterior edentulous region using CBCT data of a large patient cohort, and (2) compare the biomechanical behavior of implant-implant and tooth-implant supported fixed partial denture designs under different oblique loading conditions using three-dimensional finite element analysis. Methods Selection of subjects In this study, CBCT image data of 524 patients who underwent maxillofacial CBCT examination at the Affiliated Stomatological Hospital of Nanjing Medical University from January 2020 to December 2023 were collected and analyzed in accordance with the inclusion and exclusion criteria, including 275 males and 249 females. The age range was 18 to 81 years, with an average age of 53.5 years. Inclusion criteria: ① Age 18 years or older; ② The face is basically symmetrical, with no obvious deviation deformity; ③ Defect in the mandibular anterior teeth area (including but not limited to the absence of mandibular incisors), but no impacted teeth, supernumerary teeth or space-occupying lesions of the jawbone in the missing tooth area; ④ Alveolar bone resorption in the mandibular anterior area is mainly manifested in the horizontal direction; ⑤ The CBCT images are clear, with no obvious double images or artifacts. Exclusion criteria: ① Moderate or severe crowding of the dentition with obvious torsion or misalignment; ② History of orthodontic, orthognathic treatment or oral surgery. Experimental tools Image data acquisition was performed using a cone-beam CT machine (NewTom VG, QR srl, Verona, Italy), and the collected image data were stored in Digital Imaging and Communications in Medicine (DICOM) format and imported into software Mimics 21.0 (Material, Belgium) for three-dimensional image reconstruction and data measurement. Using computer-aided design software 3Shape Dental System 2020 (3Shape, Denmark) and reverse engineering software Geomagic Wrap 2021 (3D Systems, USA), Finite element analysis was performed using 3D modeling software SolidWorks 2023 (SolidWorks, USA) and finite element analysis processing software Ansys 2023 R1 (Ansys, USA). Image measurement In this study, the Federation Dentaire Internationale tooth position representation method was adopted (for example, the left lower central incisor was represented by 31). Bone tissue (different parts of the alveolar ridge) around the mandibular anterior tooth loss site was identified in the three-dimensional model, and the CBCT of the study subjects was calibrated from the coronal, sagittal, and horizontal planes. Due to the absence of the patient’s mandibular anterior teeth, the position of the mandibular midline must be determined first. In CBCT, the position of the mandibular midline can be determined by combining coronal and horizontal positions. First, observe the arc-shaped contour of the alveolar ridge in the horizontal position, find the symmetrical highest points on both sides, use tools to mark the highest points on both sides, and draw a horizontal line connecting the two points. The perpendicular bisector of this horizontal line is the candidate position for the midline (as shown in Fig. 1A). Verify the position of the center line in the coronal view to ensure symmetry (as shown in Fig. 1B). Mark the predicted sites of the central incisors (about 3–5 mm beside the midline, 4 mm in this study) and the lateral incisors (about 6–8 mm beside the midline, 7 mm in this study) in the horizontal view. After marking (as shown in Fig. 1C), confirm the sites in the sagittal position and measure the alveolar bone width at the sagittal marking points (as shown in Fig. 1D). The following positions were selected for measurement: 1 mm, 3 mm, 5 mm (as shown in Fig. 1E) at the top root of the alveolar ridge of 31, 32, 41, 42. All imaging measurements were performed by the same examiner. To assess measurement reproducibility, 10% of the total cases were randomly selected for repeated measurement after an interval of at least two weeks. Fig. 1CBCT measurement procedure for the mandibular anterior edentulous region. A Determination of the mandibular midline on the axial plane. B Confirmation of the mandibular midline on the coronal plane. C Identification of the reference points corresponding to the four mandibular incisor positions. D Sagittal views of the alveolar bone at different mandibular incisor sites. E Measurement of buccolingual alveolar bone width on the mid-sagittal plane. Bone width was measured at three horizontal levels perpendicular to the long axis of the tooth: 1 mm, 3 mm, and 5 mm apical to the alveolar ridge crest (ARC) CBCT measurement procedure for the mandibular anterior edentulous region. A Determination of the mandibular midline on the axial plane. B Confirmation of the mandibular midline on the coronal plane. C Identification of the reference points corresponding to the four mandibular incisor positions. D Sagittal views of the alveolar bone at different mandibular incisor sites. E Measurement of buccolingual alveolar bone width on the mid-sagittal plane. Bone width was measured at three horizontal levels perpendicular to the long axis of the tooth: 1 mm, 3 mm, and 5 mm apical to the alveolar ridge crest (ARC) Finite element model building The process commences with the importation of selected CBCT image data into Mimics Medical, a specialized medical image editing and processing software, utilizing the DICOM format. Threshold adjustments are made to establish a mask, followed by the application of the region growth function to preliminarily segment the desired jawbone regions. A layer-by-layer editing approach is adopted to refine the mask and normalize the bone marrow cavity, thereby isolating the alveolar bone segments of the upper and lower incisors. Upon verification of the model’s completeness and accuracy, noise removal and a preliminary smoothing process are conducted, followed by exportation in STL file format. The file is subsequently imported into Geomagic Wrap, a reverse engineering software, for point cloud data processing. The outer surface is optimized through the redraw mesh function, and the model is further refined using tools such as nail removal, feature removal, noise removal, relaxation, and quick smoothing, culminating in the development of a preliminary cortical bone model. The jawbone model is uniformly offset inward by 1 mm using the offset function, thereby generating the cancellous bone model. In the precise surface module, a sequential execution of contour line generation, surface sheet construction, grid construction, and surface fitting operations is performed on the jawbone model to achieve the ideal NURBS surface. Three-dimensional solid models of cortical and cancellous bone for the upper and lower incisors are constructed and exported in STP file format. For generation of the periodontal ligament, a uniform 0.25 mm layer was created around the selected tooth model using the offset/thickening method [16–18]. The 3Shape D2000 optical scanner is utilized to scan the Straumannϕ3.3 mm * 10 mm bone horizontal cone-columnar implant attached to the Straumannϕ4.8 mm * 5 mm narrow neck healing abutment, capturing the implant model data of the attached abutment and integrating it with the mandibular model. The implant model data in STL format is imported into Geomagic Wrap, optimized as previously described, and exported as a three-dimensional solid model in STP format. The process involves importing the data pertaining to the mandibular cortical and cancellous bones, as well as the implant models with attached abutments, into SolidWorks for assembly in accordance with the defined experimental groups. The implant placement and axial direction are aligned with the original natural teeth, exhibiting a slight overall deviation towards the lingual side, ensuring that the long axis of the implant remains essentially parallel to that of the restoration, while maintaining sufficient bone volume on the labial side. Subsequently, the two sets of jawbone models, incorporating the implanted implants, are imported into the 3Shape Dental System software in STL format. Restoration models are then designed, taking into account standard morphological criteria and the position of the opposing teeth. These restoration models are imported into the Geomagic reverse engineering software in STL file format for optimization as previously outlined, followed by exportation as three-dimensional solid models in STP file format. The final step entails assembling the cortical bone, cancellous bone, periodontal ligament, implants with attached abutments, and restorations within SolidWorks, thereby completing the construction of the solid model. Based on previous finite element studies, all materials were assumed to be homogeneous, isotropic, and linearly elastic [13, 14]. The Young’s modulus and Poisson’s ratio assigned to each material are summarized in Table 1 [13]. Prior to the final analysis, iterative mesh refinement was performed with particular attention to the implant-bone interface, which represented the primary biomechanical region of interest in this study. Refinement was continued until the change in model stress between two consecutive mesh refinements was less than 5%. Based on this procedure, the global mesh size was set to 0.3 mm. Local refinement was applied at the implant-bone interface, where the mesh size was reduced to 0.1 mm. In cortical and cancellous bone regions without direct implant contact, the mesh size was moderately enlarged, with a maximum element size of 1.5 mm. Table 1Material properties used in the finite element modelMaterialYoung’s modulus (MPa)Poisson’s ratioCortical bone [13]13,7000.30Cancellous bone [13]1,3700.30Periodontal ligament [13]690.45Titanium (implant/abutment) [19]105,0000.37Zirconia crown [20]210,0000.31Dentin [21]186,0000.31 Table 1Material properties used in the finite element modelMaterialYoung’s modulus (MPa)Poisson’s ratioCortical bone [13]13,7000.30Cancellous bone [13]1,3700.30Periodontal ligament [13]690.45Titanium (implant/abutment) [19]105,0000.37Zirconia crown [20]210,0000.31Dentin [21]186,0000.31 Material properties used in the finite element model Complete osseointegration was assumed between the implant and the surrounding bone, and all relevant interfaces were defined as bonded contacts with no relative movement [14, 22, 23]. Accordingly, the contacts between the implant and cortical bone, the implant and cancellous bone, the cortical and cancellous bone, and the abutment and prostheses were all assigned as bonded contacts. This assumption was adopted to represent an idealized delayed-loading condition and to enable controlled comparison between restorative designs. Fixed constraints were applied to the proximal, distal, and middle sections of the mandibular incisors [24]. The experimental setup involves the use of static loading, with independent coordinate systems established for each tooth position within the crown restoration. A static load of 50 N per tooth position was applied to the incisal edge of the mandibular incisor restorations to approximate anterior functional loading under controlled conditions (corresponding to a loading area of approximately 1mm2) [25–28]. To simulate different oblique loading scenarios related to anterior overjet, the loading direction was applied from the labial side toward the lingual side at 30°, 45°, and 60° relative to the long axis of the implant restoration [9, 10] (Fig. 2A). This loading strategy was intended for comparative biomechanical analysis and does not reproduce the full complexity of in vivo mastication or parafunctional loading. Fig. 2Loading conditions and finite element models. A Schematic illustration of the three oblique loading directions applied to the incisal edge of the mandibular incisor restoration. Loads were applied at angles of 30°, 45°, and 60° relative to the long axis of the restoration to simulate different overjet-related loading conditions. The yellow coordinate system indicates the local reference axis for load application. B Upper: Schematic representations of the two restorative configurations evaluated in this study: Model I, an implant-implant supported fixed partial denture (33I–42I), and Model II, a tooth-implant supported fixed partial denture (33I–43T). I = implant; T = natural tooth. Lower: Corresponding preliminary three-dimensional finite element models of the two restorative configurations Loading conditions and finite element models. A Schematic illustration of the three oblique loading directions applied to the incisal edge of the mandibular incisor restoration. Loads were applied at angles of 30°, 45°, and 60° relative to the long axis of the restoration to simulate different overjet-related loading conditions. The yellow coordinate system indicates the local reference axis for load application. B Upper: Schematic representations of the two restorative configurations evaluated in this study: Model I, an implant-implant supported fixed partial denture (33I–42I), and Model II, a tooth-implant supported fixed partial denture (33I–43T). I = implant; T = natural tooth. Lower: Corresponding preliminary three-dimensional finite element models of the two restorative configurations The finite element model group under consideration involves models based on the absence of mandibular teeth 33 through 42, with implants of 3.3 mm in diameter and 10 mm in length. Model I consists of two implants, strategically placed at the left lower canine (tooth 33) and right lower incisor (tooth 42) sites. These implants are subsequently interconnected via a double-ended fixed bridge, extending from tooth 33 to tooth 42, denoted as 33I-42I. Model II, on the other hand, features a solitary implant at the left lower canine (tooth 33) site, which is then connected to the right lower natural canine. This configuration is complemented by a double-ended fixed bridge that spans the gap from tooth 33 to tooth 43, identified as 33I-43T. (Fig. 2B) CBCT imaging statistical analysis Statistical analysis of the measurements was performed using SPSS 27.0 software. Kolmogorov-Smirnov test was used to determine that all measurements were in the normal distribution. Descriptive statistical analysis was used to evaluate the alveolar bone thickness at 1 mm, 3 mm, 5 mm apical to the crest in the mandibular incisor area (31, 32, 41, 42) following tooth loss. The differences between genders were analyzed using the independent sample t-test, and the analysis results were expressed as mean ± standard deviation. The significance level was set at 95%, meaning the difference was considered statistically significant when P < 0.05. All imaging data were measured by the same examiner, and 10% of the total number of cases were randomly selected for repeated measurement at intervals of at least two weeks. Finite element computational analysis In the finite element analysis software Ansys 2023, multiple loading methods were adopted to solve and calculate the above groups of models respectively. For metallic components such as implants and abutments, stress distribution was evaluated using von Mises equivalent stress [29]. The implant and abutment components were modeled as titanium alloy, and previous reports indicate that commonly used Ti-6Al-4 V alloy exhibits a yield strength on the order of approximately 800–850 MPa [30]. For the surrounding bone and tooth-related structures, the results were interpreted comparatively to identify stress concentration patterns among different restorative configurations and loading conditions. Because the present model did not simulate fracture, fatigue behavior, bone remodeling, or long-term biological adaptation, the stress values were not used to infer absolute clinical safety thresholds. Results Gender differences The gender composition and average age in the sample data are shown in Fig. 3A. CBCT images of 524 patients were collected in accordance with inclusion and exclusion criteria in this study. Among the images included in the study, 275 were male, accounting for 52.5%, with an average age of (49.559 ± 15.463) years; There were 249 females, 47.5%, with an average age of (52.013 ± 14.848) years. The age distribution of the sample data was largely similar between different genders. Fig. 3B shows the measurements of bone width at three depths of 1 mm, 3 mm, and 5 mm apical to the crest in the four tooth positions 31, 32, 41, and 42. This figure presents the descriptive statistical analysis results of the sample in this study, covering mean, standard deviation, minimum and maximum values at different measurement positions. The distribution of the data is wide, and the differences from minimum to maximum are significant. Fig. 3Sample characteristics and sex-related differences in alveolar bone width. A Sex composition and mean age of the study sample. B Descriptive statistics of alveolar bone width at different mandibular incisor sites and measurement depths. C Comparison of alveolar bone width between female and male patients at different sites and depths. (*, P < 0.05; **, P < 0.005) Sample characteristics and sex-related differences in alveolar bone width. A Sex composition and mean age of the study sample. B Descriptive statistics of alveolar bone width at different mandibular incisor sites and measurement depths. C Comparison of alveolar bone width between female and male patients at different sites and depths. (*, P < 0.05; **, P < 0.005) Figure 3C presents a comparison of all measurement metrics between females (n = 249) and males (n = 275). The results indicate that males generally exhibited greater residual bone width than females, with statistically significant gender differences observed at all measurement sites (P < 0.05). Particularly notable differences (marked as **) were found at 31 (1 mm), 41 (3 mm, 5 mm), and 42 (5 mm). The largest absolute t-values were observed at 41 (3 mm: t = -4.785; 5 mm: t = -4.751), indicating the strongest effect of gender difference. Although 42 (1 mm) also showed a statistically significant difference (P = 0.042), the relatively small t-value (t = -2.034) suggests that the practical impact of gender variation at this specific location may be limited. Correlations with age This study further used One-way ANOVA to compare the measurements among the groups of 18–30 years (n = 38), 30–50 years (n = 199), and over 50 years (n = 287), as shown in Fig. 4A. All indicators showed significant differences among different age groups (P < 0.01), suggesting that age may have a certain degree of influence on these measurements. Fig. 4Age-related differences and correlations in alveolar bone width. A Comparison of alveolar bone width among different age groups (18–30 years, 30–50 years, and > 50 years) using one-way ANOVA. B Correlation between alveolar bone width and age at different mandibular incisor sites and measurement depths. (**, P < 0.005) Age-related differences and correlations in alveolar bone width. A Comparison of alveolar bone width among different age groups (18–30 years, 30–50 years, and > 50 years) using one-way ANOVA. B Correlation between alveolar bone width and age at different mandibular incisor sites and measurement depths. (**, P < 0.005) For most tooth positions and measurement depths, the average available bone width obtained from the measurements showed a downward trend with age. For example, at measurement depths of 1 mm, 3 mm, and 5 mm, the average values in the over 50 group were significantly lower than those in the 18–30 group. 32 also showed a similar age-related downtrend at 1 mm and 5 mm, but at 3 mm, the average for the 30–50 group was slightly higher than the other two groups. P values for all tooth positions and measurement depths were less than 0.05, indicating statistically significant differences in measurement depths among different age groups. At a depth of 5 mm, F = 62.562, with a larger F value indicating a stronger effect of the difference between groups. For 32, at a depth of 3 mm, the 30–50 age group had the highest average value, which may be related to sample characteristics or other factors and requires further study. To further explore the relationship between bone width and age, Pearson correlation coefficients were used for analysis in this study, and the results are shown in Fig. 4B. Overall, there was a significant negative correlation between bone width and age, and the correlation coefficient ranged from − 0.09 to -0.4, indicating a consistent, albeit modest, decline in alveolar bone dimensions with advancing age. Bone width at the 31(5 mm) site had the highest correlation coefficient with age (r = -0.473, P < 0.01), suggesting that this indicator decreased most significantly with age. The correlation coefficient for 31(3 mm) was also relatively high (r = -0.362, P < 0.01). In contrast, 32(3 mm) had the lowest correlation coefficient (r = -0.091, P < 0.05), suggesting that although its association with age was statistically significant, it was relatively weak. In addition, other measurement points, such as 41 (3 mm) and 41 (5 mm), also showed significant negative correlations (r = -0.288 and − 0.293, P < 0.01), suggesting that the trend of decreasing values at this location with age is also worthy of attention. To sum up, available bone width indicators at different sites were negatively correlated with age to some extent, but with varying intensities. Implant stress distribution under different overjet Clinically, for patients with mandibular anterior tooth loss, a double-ended fixed bridge is often chosen to restore the dentition defect of the anterior teeth, that is, two implants are placed at both ends of the bridge to form an implant-supported double-ended fixed bridge. Therefore, this study constructs a double-ended fixed bridge (33I–42I) supported by implants in the left mandibular canine and the right mandibular incisor, referred to as Model I. Below are the von Mises equivalent stress contour plots of the implant (Fig. 5A, upper part) and the alveolar bone (Fig. 5C, upper part) under different occlusal loading conditions in Model I, along with comparisons of the maximum von Mises equivalent stress (Fig. 5B and D).It can be seen from Fig. 5A (upper part) that the maximum stress is located at the cervical part of the tooth, particularly on the labial side at the junction of the 42 implant and abutment. As shown in Fig. 5B, the maximum stress value in Model I is 281.01 MPa when the loading angle is 30°. When the loading angle increases to 45°, the maximum stress value reaches 318.52 MPa, and at 60°, it increases further to 334.36 MPa. Fig. 5Stress distribution in implants and surrounding alveolar bone under different oblique loading conditions. A von Mises stress contours of the implant components in Model I and Model II under 30°, 45°, and 60° loading at the incisal edge. In both models, stress concentration was mainly observed at the implant-abutment neck. B Comparison of the maximum implant von Mises stress values under different loading conditions in Model I and Model II. C von Mises stress contours of the surrounding alveolar bone in Model I and Model II under 30°, 45°, and 60° loading at the incisal edge. Stress concentration was mainly observed in the crestal cortical bone adjacent to the implant neck. D Comparison of the maximum cortical bone stress values under different loading conditions in Model I and Model II Stress distribution in implants and surrounding alveolar bone under different oblique loading conditions. A von Mises stress contours of the implant components in Model I and Model II under 30°, 45°, and 60° loading at the incisal edge. In both models, stress concentration was mainly observed at the implant-abutment neck. B Comparison of the maximum implant von Mises stress values under different loading conditions in Model I and Model II. C von Mises stress contours of the surrounding alveolar bone in Model I and Model II under 30°, 45°, and 60° loading at the incisal edge. Stress concentration was mainly observed in the crestal cortical bone adjacent to the implant neck. D Comparison of the maximum cortical bone stress values under different loading conditions in Model I and Model II Model II (33I–43T) was established to represent a tooth-implant supported restorative configuration incorporating the natural right mandibular canine (43). The von Mises stress contours and maximum implant stress values under different loading conditions are shown in Fig. 5A (lower panel) and Fig. 5B, respectively. To investigate whether natural teeth can be combined with implants to repair mandibular anterior dentition defects, we developed Model II (33I–43T), a fixed bridge that strategically incorporates the natural right mandibular canine (43). The color cloud of the von Mises equivalent stress of the implant and the comparison of the maximum von Mises equivalent stress of Model II under different occlusal loading methods are shown in Fig. 5A (lower part) and Fig. 5B, respectively. It can be seen from the bar chart that the maximum von Mises stress in Model II was 306.31 MPa at a 30° loading angle, increasing to 348.29 MPa at 45°, and reaching 362.78 MPa at 60°.The above research results indicate that the von Mises equivalent stress value of the implant gradually increases with the increase of the loading angle, suggesting the greater the overjet, the greater the stress on the implant. Comparative analysis of cortical bone stress on implants under different overjet The maximum stress in the alveolar bone was concentrated in the cortical bone surrounding the implant sites, as shown in Fig. 5C. In Model I, the maximum stress values were 106.20 MPa at 30°, 124.53 MPa at 45°, and 135.73 MPa at 60°. In Model II, the highest stress occurred around the 33 implant site, with values of 154.62 MPa at 30°, 173.26 MPa at 45°, and 182.52 MPa at 60°(Fig. 5D). In both models, cortical bone stress increased with increasing loading angle, and Model II showed higher peak cortical bone stress around the implant site than Model I under the same loading conditions. These findings indicate that both loading direction and restorative configuration substantially influence stress concentration in the peri-implant cortical bone. However, these results should be interpreted as comparative biomechanical observations within the present simplified finite element framework rather than as direct predictions of clinical bone resorption or tissue damage. Discussion Mandibular anterior tooth loss, resulting from caries, trauma, or periodontal diseases, commonly impairs both masticatory efficiency and facial aesthetics [31]. Following tooth loss, bite forces shift posteriorly, depriving anterior regions of essential mechanical stimulation and accelerating horizontal/vertical alveolar bone resorption [32]. In cases of advanced periodontitis, residual bone volume diminishes further post-extraction [33]. Notably, Jensen’s findings indicate that following implant placement in the anterior mandible, labial bone wall thickness below 1.5 mm correlates with substantial vertical bone loss [34, 35]. Maintaining 1.5 ~ 2 mm labial bone proves critical for long-term implant stability [35, 36]. Consequently, ideal surgical conditions require ≥ 6 mm alveolar bone thickness at implant sites as in clinical implant practice, the minimum diameter of implants used in the anterior region typically ranges from 3.3 mm to 3.5 mm [37]. However, our measurements revealed mean bone widths < 6 mm across key locations (teeth 31, 32, 41, 42 at 1 mm/3 mm depths; some teeth at 5 mm depth), demonstrating severe bone deficiency, necessitating prior augmentation before implant placement. Statistical analysis revealed significantly greater alveolar bone width in males versus females across all tooth positions and measurement depths (P < 0.05). Notably, pronounced gender disparities emerged at both 3 mm and 5 mm depths within the 41 site (t = − 4.785 and t = − 4.751; P < 0.001), demonstrating superior osseous dimensions in males. Consequently, meticulous evaluation of bone quantity becomes imperative for female patients undergoing implant restoration or augmentation procedures to mitigate risks of bone dehiscence or soft tissue recession following implant placement. Age-stratified comparisons evidenced progressive reduction in bone width concomitant with advancing age. Mean values measured at the 31 site (1 mm, 3 mm, 5 mm depths) exhibited statistically significant decrements in cohorts exceeding 50 years relative to those aged 18–30 (P < 0.001). Pearson correlation substantiated a robust negative association between age and bone width (r: −0.091 to − 0.473; P < 0.05), most markedly at the 5 mm depth of tooth 31 (r = − 0.473). Thus, the bone volume of the elderly warrants heightened clinical vigilance. Variations in bone width across measurement depths underscore the necessity of depth-specific implant sizing to optimize primary stability and long-term osteointegration success [38]. These findings emphasize the critical role of personalized pretherapeutic assessment for treatment planning accuracy [3]. The insufficient soft and hard tissue support commonly encountered after mandibular anterior tooth loss makes implant surgery in this region highly technique-sensitive [39]. Although bone augmentation may facilitate implant placement, it is often associated with greater surgical trauma, a longer treatment course, and an increased risk of complications, which may limit its applicability, particularly in elderly patients or those with poor systemic conditions [40, 41]. Under conditions of limited anterior alveolar bone volume, optimizing restorative design to improve mechanical distribution while reducing reliance on augmentation procedures becomes an important issue. In selected cases, a tooth-implant supported restoration may be considered as a possible alternative, as previous studies have suggested that combining natural teeth with a limited number of implants may offer certain biomechanical and functional advantages, including splinting effects, additional retention and support, protection against tooth movement, and transmission of occlusal forces more favorably along the long axis of the abutment teeth [42]. Therefore, fixed partial dentures supported by natural teeth and implants may represent an optional treatment strategy when indications are carefully controlled. In this context, finite element analysis provides a useful method for comparing stress distribution patterns among different restorative designs under standardized loading conditions. Accordingly, the present study used three-dimensional finite element analysis to compare the biomechanical behavior of implant-implant supported and tooth-implant supported restorations in the mandibular anterior region, rather than to directly predict clinical safety, tissue response, or long-term prognosis. Several limitations of the present finite element model should be acknowledged. First, all materials were assumed to be homogeneous, isotropic, and linearly elastic, which does not fully reflect the anisotropic and heterogeneous nature of biological tissues. Second, complete osseointegration was assumed at the implant-bone interface by using bonded contact conditions, representing an idealized delayed-loading scenario and not accounting for partial osseointegration or interfacial micromotion. Third, only static oblique loading was applied. Therefore, dynamic mastication, cyclic fatigue, parafunctional loading, and time-dependent periodontal ligament behavior were not simulated. In addition, screw preload, preload loss over time, and bone remodeling were not included in the present model. For the CBCT morphometric analysis, repeated measurements were performed in 10% of the cases after an interval of at least two weeks. However, formal intra-observer reliability coefficients were not calculated in the present study and should be considered in future work. These simplifications limit direct clinical extrapolation. However, they still allow controlled comparative analysis of different restorative designs under standardized conditions, which was the primary objective of this study. This study evaluated the von Mises equivalent stress of implant selection, which is also the most widely used indicator for the mechanical evaluation of implants in current research [43]. The stress in bone tissue is mainly concentrated at the alveolar ridge at the implant-bone interface, and the stress on cortical bone is much greater than that on cancellous bone [44]. Therefore, this study focuses on the stress distribution of cortical bone in alveolar bone. von Mises equivalent stress was used for both natural teeth and periodontal ligament [45, 46]. By observing the stress distribution cloud maps of the two models under different loading conditions, it can be found that: The maximum von Mises stress at the implant-bone interface was concentrated on the alveolar bone at the neck of the implant, which is consistent with the fact that the resorption of peri-implant bone tissue mainly occurs at the cortical bone at the neck in clinical practice. In both restorative configurations, the highest stresses were concentrated mainly at the implant-abutment neck and the adjacent crestal cortical bone. This pattern is consistent with the view that the implant neck region represents a mechanically critical area in peri-implant biomechanics [9, 10]. Nevertheless, the present study did not simulate bone remodeling, fatigue, or long-term biological adaptation; therefore, no absolute biological threshold or direct clinical consequence should be inferred from these stress values. Given that overjet is closely associated with occlusal trauma [47] and that the deepening of overjet is typically accompanied by an increase in the occlusal angle [48], this study simulates three different overjet relationships: shallow, moderate and deep overjet using three loading angles (30°, 45°, and 60°). When the direction of the occlusal load changed, the stress on the alveolar bone also changed [49]. In this study, for the mandibular incisors, the loading site was set at the cut edge of the crown restoration, and the loading direction was oblique from the labial side to the lingual side, respectively at 30°, 45°, 60° with the long axis of the implant restoration as different overjet relationships, with a total of three loading methods. Under oblique loading, with the increase of loading angle, the maximum von Mises stress in each model increased significantly, suggesting that with the increase of overjet depth, stress concentration in the neck is more likely to occur. This result suggests that the occlusal relationship of the patient is an important factor affecting the stress distribution when choosing TISP [50]. In all models, whether it was simple implant support or combined natural tooth-implant support, the maximum stress was concentrated on the neck of the implant (especially the labial side) and the corresponding alveolar cortical area, which was consistent with the characteristics of the rigid connection at the implant-bone interface [14]. The use of tooth-implant supported prostheses remains controversial. Previous studies have reported potential biological and technical complications related to the biomechanical mismatch between natural teeth and implants, including nonuniform stress transfer, marginal bone loss, prosthetic complications, and possible intrusion of the natural abutment tooth in certain designs [5, 6]. At the same time, some studies have suggested that acceptable outcomes may be achieved in carefully selected cases under controlled indications [51]. Therefore, the present finite element findings should be interpreted within this broader context of ongoing clinical debate rather than as support for routine clinical use of TISP. Conclusions Within the assumptions of this simplified three-dimensional finite element model, mandibular anterior tooth-implant supported and implant-implant supported restorations exhibited different stress distribution patterns under oblique loading. Stress concentrations were mainly observed at the implant-abutment neck and the adjacent crestal cortical bone, and greater loading angles were associated with increased stress levels. These findings provide comparative biomechanical information that may assist restorative planning in selected mandibular anterior edentulous cases. However, because the present model was based on simplified material assumptions, bonded implant-bone contact, and static loading, the results should not be interpreted as direct evidence of clinical safety, tissue response, or long-term prognosis. Further experimental and long-term clinical studies are needed to validate the clinical relevance of these observations.