Biomechanical Evaluation of Different Implant-Abutment Connections, Retention Systems, and Restorative Materials in the Implant-Supported Single Crowns Using 3D Finite Element Analysis

Replacing missing teeth with implant-supported prostheses has been deemed as a highly effective treatment option that has demonstrated high success rates, regardless of the type of rehabilitative treatment.¹  With increased survival of dental implants, maintenance of the peri-implant bone is considered an important factor in the long-term success of implant rehabilitation from both a functional and aesthetic perspective.² 

Therefore, biomechanical studies evaluating the behavior of variables related to implants and prosthesis that contribute to increased stress have become increasingly important.^3,4  Rehabilitation success is directly related to the appropriate processing of stress distributions because of the contact between the implant and bone during mastication.⁵  Therefore, it is important to avoid occlusal overloading that can compromise the longevity of the rehabilitative treatment.⁶ 

Among the different factors that influence stress distribution, the implant-abutment interface is an important biomechanical factor. Biomechanical studies have shown inferior biomechanical behavior for implants with an external connection as compared to those with an internal connection. This is especially true for Morse taper implants.^3,7  However, some studies have reported no difference in the biomechanical behavior between internal and external connections.^8,9 

Retention systems can also affect the stress distribution. Such systems are usually either screw or cement retained. The choice between the 2 is often based on the experience and clinical preference of the restoring clinician.²  Biomechanical studies have reported the superiority of cemented prostheses.^10,11  However, there is no consensus on this, as some studies have reported no difference between the 2 retention systems.^12,13 

The type of restorative material used can also influence stress distribution in the structures.¹⁴  For many years, metal-ceramic prostheses have been considered by some clinicians to be the gold standard of rehabilitation with an implant-supported prosthesis. However, currently, there is increasing demand for metal-free restorations as a more esthetic rehabilitative treatment option.¹⁵  Zirconia possesses mechanical strength similar to that of metals,¹⁶  especially in monolithic restorations with no ceramic layers that may increase the incidence of complications like ceramic chipping.¹⁷  However, there is little evidence in the existing literature regarding the biomechanical influence of restorative materials.

Some biomechanical studies have evaluated the influence of these variables. However, the interaction between these variables has not yet been evaluated. Therefore, the aim of this in silico study was to evaluate the biomechanical behavior of the bone, as well as the implants and their associated components. Our null hypotheses was as follows: (1) the implant-abutment interface would not interfere with the biomechanical behavior of the structures, (2) cement-retained and screw-retained systems would have similar biomechanical behavior, and (3) different restorative materials would not influence the stress distribution of the evaluated structures.

Three variable factors were considered: the implant-abutment interface (external hexagon and Morse taper implants), retention systems (cement- and screw-retained prostheses), and the restorative material of the crown (metal-ceramic and single monolithic crowns). Eight models were used to verify the biomechanical influence of these variables (Table 1).

Each model was simulated to represent a bone section for the maxillary first molar region, with bone type IV consisting of a 1-mm cortical layer and low-density trabecular bone as per Lekholm and Zarb.¹⁸  The InVesalius software (Centro de Tecnologia de Informação Renato Archer, Campinas SP, Brazil) was used for bone modeling. Next, surface simplification was performed using the Rhinoceros software (NURBS Modeling, Seattle, Wash).

Each simulated model contained an implant (external hexagon and Morse taper implants with Ø 4 mm × 10 mm length) with an abutment (universal castable long abutment [UCLA]) that supported a metal–ceramic (with feldspathic porcelain) or zirconia monolithic single crown with a screw- or cement-retained system. The design of the structures was obtained from an original version (Conexão Sistemas de Prótese Ltda, Aruja SP, Brazil) and was simplified using the SolidWorks (SolidWorks Corporation, Concord, Mass) and Rhinoceros softwares. To eliminate an additional biomechanical variable (macrogeometry), the external hexagon and Morse taper dental implants were simulated with the same macrogeometry to isolate only the biomechanical influence of the implant-abutment connection. This was done because the macrogeometry of dental implants affects the stress distribution of structures.¹⁹ 

The UCLA abutment was simulated to standardized the implant-abutment interfaces, as were the retention systems for the metal-ceramic crowns. This was done to isolate the tested variables. The crowns were simulated based on an artificial first molar, using a 3-dimensional (3D) scanner (MDX-20w, Roland DG Brasil, Cotia SP, Brazil) for scanning (Odontofix Indústria E Comércio De Material Odontológico, Ribeirão Preto SP, Brazil). For the metal-ceramic crown, nickel-chromium alloy was used for the framework, and feldspathic porcelain was used as a veneer on the external crown surfaces. The monolithic single crowns were simulated from zirconia, with the same geometry as that used for the metal-ceramic crowns. However, an indexing titanium base (TiBase) was used to simulate the milled crowns for this system. In the cemented prostheses, a 50-μm cementation line was simulated.¹⁰ 

After the modeling phase, the solids were exported to the pre- and postprocessing finite element analysis (FEA) FEMAP 11.4.2 software (Siemens PLM Software, Santa Ana, Calif) to generate models with tetrahedral parabolic solid elements. The mechanical properties of each simulated material were determined according to previously published studies (Table 2).^3,10,11,16  All materials were considered isotropic, homogeneous, and linearly elastic.

The crown-abutment and implant-abutment connections were assumed to have symmetrical contacts. All other contacts were also assumed to be symmetrically welded. Constraint conditions were fixed for all axes (x, y, and z) at the anterior and posterior surfaces of each bone section. All other model parts were unrestricted. A force of 200 N was applied axially at 4 points on the internal slope of each cusp, and 100 N was applied obliquely (45° at the long axis of the dental implant) at 2 points on the internal slope of the buccal cusps. After the preprocessing step in the FEMAP 11.4.2 software, the models were exported to the NEI Nastran 11.0 software package to undergo mathematical calculations (Noran Engineering, Westminster, Calif). The models were then imported to FEMAP 11.4.2 for postprocessing and visualization of the stress maps. von Mises stress maps were used to evaluate the biomechanical behavior of the implants and their components in meagapascals (MPa). The microstrain (με) was used to assess the strain deformation of the cortical bone. A bar chart with the 50 highest extracted values from all elements of the cortical bone and fixation screw for each simulated model was also used.

The Morse taper implants had lower microstrain values (range, 1295–1432 με) on the cortical bone tissue under axial loading compared with the external hexagon implants (range, 1832–2715 με). However, no difference was observed in the microstrain between the retention systems (cement retained: range, 1352–2715 με; screw retained: range, 1295–2544 με) and the restorative material crowns (metal-ceramic: range, 1295–2544 με; monolithic: range, 1297–2714 με; Figures 1 and 2). Higher microstrains were observed during oblique loading. Morse taper implants displayed better biomechanical behavior in terms of microstrain, mainly in the buccal area (range, 3862–4598 με) compared with external hexagon implants (range, 11 757–13 587 με). This result was independent of the retention system (cement retained: range, 3862–13 128 με; screw retained: range, 4240–13 587 με) and restorative material crowns (metal-ceramic: range, 3862–13 587 με; monolithic: range, 4029–13 566 με; Figures 1 and 2).

Under axial loading, high stress was concentrated in the cervical or distal wall of the dental implant, abutment, and fixation screw. External hexagon implants displayed higher stress values at the center of the fixation screw using a cement-retained prosthesis (range, 66–75 MPa) compared with a screw-retained prosthesis (range, 22–28 MPa). Compared with the external connection, the screw-retained prostheses of the Morse taper implants had higher stress in the fixation screw (range, 36–40 MPa) than the cement-retained prostheses (range, 30–35 MPa), but both stresses were considered lower than those of external hexagon cemented prosthesis. The metal-ceramic (range, 23–74 MPa) or monolithic (range, 22–75 MPa) crowns did not influence the stress distribution in the fixation screw and the implant-associated components (Figures 3 and 4).

In the oblique loading, as observed previously, the external hexagon with cemented prosthesis displayed higher stress on the fixation screw (range, 929–1160 MPa) compared with a screwed prosthesis (range, 620–763 MPa). Furthermore, in both retention systems for external hexagon implants, the stress values in the fixation screws were higher than those for Morse taper implants with screw-retained (range, 61–71 MPa) and cement-retained (range, 56–60 MPa) prostheses. However, no difference in stress was observed between different retention systems (screw or cement retained) with respect to Morse taper implants. Furthermore, there was no difference in stress distribution for the different restorative materials, independent of the connection and retention system (metal-ceramic: range, 63–1159 MPa; monolithic: range, 61–1170 MPa; Figures 3 and 4).

The null hypothesis that the implant-abutment interface would not affect the microstrain on the bone or stress distribution in the implants and components was rejected. Morse taper implants displayed better biomechanical behavior in the bone tissue compared with the external hexagon implants. These results are consistent with the previously published studies.^3,21  This difference in biomechanical behavior can be related to Morse taper implants having a more stable connection because of an internal conical connection and having mechanical friction between the abutment and implant inner wall.²¹  Additionally, the centralization of stress along the long axis of the implant may also assist in reducing the stress in the cortical bone region.³  A recent meta-analysis reported that the internal connection implants (mainly those with a conical connection) afford greater bone preservation than external connection implants (external hexagon).²²  This finding is in line with the biomechanical behavior observed in this study.

These advantages are generally attributed to the platform switching concept, in which the implant-abutment connection interface remains at a distance from the bone crest, contributing to the biomechanical characteristics.^23,24  External hexagon implants might allow micromovements of the abutment, causing instability of the interface,^23,25  because the screw alone is responsible to fixate the abutment on the implant, whereas Morse taper implants provide a lock resistance to the lateral loadings because of the taper interface which prevents the movement of the abutment.²⁶  For these design reasons, the external hexagon creates increased stress between the implant and abutment adjacent to the bone.²²  The fixation screw of the external hexagon is the least resistant component of the implant-abutment connection.³  In this study, external hexagon implants displayed higher stress in the fixation screw than Morse taper implants. This can be clinically correlated because an external connection in single crowns is related to a higher incidence of screw abutment loosening or fracture.²⁷ 

No difference was observed in the microstrain values of the bone tissues with regard to retention systems. However, a significant difference was observed in the stress of the implant (mainly in the fixation screw), which rejects our second hypothesis. The different abutment types (external and internal connection) can affect stress distribution, and the similarity in the microstrain values of the cortical bone tissue can be related to the standardization of the UCLA and crown design for both retention systems.²⁸  However, there was a higher concentration of stress in the fixation screw for the cemented prostheses in the external hexagon implants, independent of the restorative materials used.

The higher stress in the fixation screw of cement-retained prosthesis concurs with the report of an increased incidence of abutment loosening for cement-retained prosthesis in a previous systematic review.²⁹  A probable explanation for this result can be associated with the difference in stress distribution between the segmented and nonsegmented prostheses³⁰ ; screw-retained prosthesis tends to transmit stress to the apical portion of the implant, whereas cement-retained prosthesis tends to concentrate the stress in the coronal portion of the implant.³¹  Thus, because the implant-abutment interface of the external hexagon implants is located in this region, this may justify the increased stress in the fixation screw of these models. Therefore, the combination of external hexagon implant and cement-retained prosthesis should be avoided. This avoidance is particularly important in situations with high masticatory loads. If an abutment screw should loosen with a cement retained crown it makes for a difficult clinical situation. The crown would have to be compromised with an access whole to retighten the loose abutment screw.³²  However, these results should be interpreted with prudence because in this FEA study, all models were simulated without a marginal misfit. The increase in the marginal gap could contribute to increased stress in the components,³³  which could interfere with the biomechanical evaluation of screw- and cement-retained prostheses. It should be noted that some studies reported that cement-retained prosthesis can exhibit better stress distribution as the cement layer may fill in framework interfacial discrepancies and assist in the equitable load distribution.^13,34 

In the literature, few studies have evaluated the biomechanical behavior of monolithic implant-supported single crowns and the influence of isolated variables. To the best of our knowledge, this is the first biomechanical study that evaluates the comparison between monolithic and metal-ceramic crowns with different implant-abutment interfaces. Metal-ceramic and zirconia monolithic implant-supported single crowns had similar biomechanical behaviors in the bone tissue, implants, and their components. These results are in agreement with those of a biomechanical study in which the effect of restorative materials on stress distribution was not observed.¹¹  The similarities between the metal-ceramic and zirconia monolithic prostheses may be attributed to similar mechanical properties (modulus of elasticity and Poisson's ratio),¹⁶  which may contribute to the sharing of stress across structures. An advantage of using monolithic restorations is the possibility of using an esthetic material without needing an interface area between the veneering ceramic and infrastructure in metal-free restorations.³⁵  This factor interferes with the biomechanisms, because the incomplete adhesion face between the ceramic layer with the structure affects the stress distribution, making the structure susceptible to failure.¹⁴  Therefore, monolithic restorations have been increasingly used to reduce these possible complications.³⁵  In addition, various restorative materials, such as lithium disilicate, zirconia-reinforced lithium silicate, and polymer-infiltrated ceramic network restorations, are considered for the monolithic implant-supported single crown rehabilitation.^36,37  Therefore, further biomechanical studies are warranted for the evaluation of as previously described materials.

The main limitations of this study are related to the possible restrictions of the computational method used in the finite element analysis. The properties of linearly elastic, isotropic, and homogeneous materials were considered, as in previous studies.³  Therefore, these results should be interpreted with caution before applying them used in clinical practice. Under the conditions of this study, using a finite element methodology has some advantages, because it allows analysis of the internal structures of the bone and the components and can provide targeted responses to the biomechanical behavior of different structures. However, more clinical studies are recommended to evaluate the influence of the different biomechanical variables that were validated by this in silico study, especially because there are no clinical studies that combine the effect of the implant-abutment connection with retention systems or restorative materials, including the newer materials available for implant-supported restorations.

Within the limitations of this study, the following conclusions can be drawn:

1. Morse taper implants can be considered as a good alternative for dental implant rehabilitation because they provided better biomechanical behavior with respect to the structures, regardless of the retention system or restorative material used.

2. Retention systems did not influence the microstrain values. However, external hexagon implants with cemented prostheses presented higher stress on the fixation screw, increasing the risk of screw loosening/fracture in the maxillary posterior area.

3. Monolithic zirconia crowns showed biomechanical behavior similar to that of metal-ceramic crowns in the cortical bone tissue and the implants and associated components.

Abbreviations

3D:

three-dimensional

FEA:

finite element analysis

TiBase:

titanium base

UCLA:

universal castable long abutment

The authors thank Renato Archer Research Center for support of the analysis and Conexão Sistemas de Prótese for support of the design of implants and components. This study was supported by Scholarship of São Paulo State Research Foundation (FAPESP: #2015/24442-8 and CNPq: 306288/2016-8).

The authors declare no conflicts of interest.