Stress distribution in prosthetic abutments: a finite element analysis comparison of conical and UCLA abutments

The effect of prosthetic abutment type on single-screwed prostheses in posterior mandibular molar rehabilitations is not yet known. Thus, the aim of this study was to evaluate the distribution of stresses in the crowns, prosthetic components, implant and bone in implant-supported restorations with or without a prosthetic abutment, maintaining an equal total height of the implant-crown set. Virtual 3-dimensional (3D) finite element models were constructed, the models were designed to represent a posterior single crown rehabilitation with a screwed retention system and external hexagon implants placed in the lower first molar region. Two rehabilitation methods were designed to simulate a monolithic zirconia crown screwed onto a conical abutment, which was screwed onto an external hexagon implant (M1); and a monolithic zirconia crown screwed directly onto the external hexagon implant using an UCLA abutment (M2). An axial load of 200 N was simulated and applied axially in the occlusal region of the restoration divided into 5 points. The quantitative and qualitative description of the maximum principal stress for crowns, von Mises stress for screws, conical abutment and implant; and minimal principal stress for cortical and medullary bone were evaluated. M1 presented similar stress distribution for crowns, cortical and medullary bone compared to M2. Conversely, the stress values were considerably higher for crowns screw and implants in the M2 group. In conclusion, single implant-supported rehabilitations of mandibular first molars using external hexagon implants presented better stress distribution on the crown screw and implants for the M1 group compared to M2.


Introduction
Single implant-supported rehabilitations have emerged as a safe and effective alternative for the treatment of partial edentulous patients (Mezzomo et al., 2014;Wang et al., 2020). This treatment presents high survival rates of 97.2% after 5 years, despite the great biomechanical challenges of this type of rehabilitation (Jung et al., 2012). Hereby, aiming to improve the biomechanics and the predictability of the single implant-supported restorations, different connection systems, abutment type and materials, surface treatments, length and diameter of abutments and implants have been proposed (Di Fiore et al., 2015;Lima de Andrade et al., 2017).
Within the connection systems, external hexagon implants developed by Bränemark have been the most used system and its retention to the prosthetic abutment occurs through a hexagon-shaped fitting. Several drawbacks are reported for this connection system and it has been suggested that, under high occlusal loads, the connection might allow micro movements, resulting in abutment screw loosening or even fatigue fracture (Dias et al., 2012;Pera et al., 2021).
It also must be considered the types of restorations that can be used with respect to the abutment type. The use of an intermediary component between the implant and prosthetic crown (conical abutment) and also prostheses screwed directly on the implant (UCLA abutment) have been used (Araújo et al., 2018). The first comprises a segmented structure with two screwed joints one at implant-abutment interface and another at abutment-prosthesis interface (Ochiai et al., 2003). For the second, a unique screw is used and screwed directly onto the implant. Implant failures are still a clinical challenge and can be associated with loading conditions, bio-tribocorrosion phenomena, biological reasons (colonization by bacteria in the internal parts of the implant and peri-implant diseases) or mechanical reasons, and micro-movement of the prosthetic components (Coelho et al., 2009;Dini et al., 2020). Thus, an adequate adjustment of the prosthetic components is necessary. Considering the fact that the implants are completely surrounded by bone tissue, and that the interface is not elastic, minimal movement can provide bone resorption and loss of components under load (Jaime et al., 2007). Additionally, higher stress on implants, screws and abutments may also predict failures (Tribst et al., 2021).
With the purpose of simulate and analyse the mechanical behavior of materials the three-dimensional (3D) finite element analysis have been extensively used. Through this in silico analysis, it is possible to simulate complex conditions without risk to patients, evaluate the performance and stress distribution in the supporting tissues with the help of a computational model (Costa et al., 2014). This method allows the individual assessment of the influence of variables on the stress generated to the peri-implant bone and on the possible deformation of the system, which is difficult to be investigated using only clinical or in vitro approaches.
In the literature, the effect of prosthetic abutment type, for single-screwed prostheses in posterior mandibular molar on zirconia crowns, crowns screws, conical abutment, abutment screws, implants, cortical and medullary bone is not yet known.
Considering that single implant-supported rehabilitation are not splinted and hence subjected to multidirectional loading that challenges the connection components and restoration structural integrity, it is important to evaluate the predictability of this type of rehabilitation (Estrela, 2018;Pereira et al., 2018). Therefore, the aim of this study was to evaluate the distribution of stresses in the crowns, prosthetic components, implant and bone in implant-supported restorations with or without a prosthetic abutment, maintaining an equal total height of the implant-crown set.

Methodology
Virtual 3-dimensional (3D) finite element models were constructed by using a modeling software program (Solid Works 2013; Dassault Systèmes Solid Works Corp). The models were designed to represent a posterior single crown rehabilitation with a screwed retention system and external hexagon implants placed in the lower first molar region. Two rehabilitation methods were designed to simulate a monolithic zirconia crown screwed onto a conical abutment, which was screwed onto an external hexagon implant (M1); and a monolithic zirconia crown screwed directly onto the external hexagon implant using an UCLA abutment (M2). The prosthetic abutment used in the M1 was a conical titanium abutment for external hexagon implants, measuring 4.1 mm in diameter and 4.5 mm in height, with 2.5 mm of metallic strap. For the group M2, an anti-rotational UCLA abutment was used and the crown was screwed directly onto the external hexagon implants. The crowns of both groups were modeled to obtain an identical occlusal anatomy, only their heights being different, since the M2 crown was screwed directly onto the implant using the UCLA abutment. Thus, the group M2 presented a greater height to compensate for the height of the conical abutment, thus maintaining an equal total height between the two group models (M1 and M2). A titanium screw was also modelled to be screwed onto the external hexagon implant measuring 3.75 mm in diameter, 11 mm in height and 4.1 mm platform. Another titanium fixation screw was modelled to screw the crown to the prosthetic abutment.
Both the implant and the cortical bone block and medullary bone were modelled and used for both M1 and M2 (Figure 1).
After modeling all components separately, they were assembled in the 3D finite element analysis software program, generating two experimental models for the groups M1 ( Figure 2A) and M2 ( Figure 2B), which were imported to the finite element analysis software (ANSYS Workbench 15.0; Ansys Inc) for mathematical analysis. A 70 µm mesh of tetrahedral elements was defined after convergence analysis of 5% and an axial load of 200 N, consistent with reports found in the literature for the posterior region ( Figure 3A) (Lemos et al., 2021;Lee et al., 2021). This load was applied axially in the occlusal region of the restoration divided into 5 points according to a normal occlusion ( Figure 3B) (Brune et al., 2019;Bordin et al., 2021).   Table 1 and the Young modulus and Poisson ratio used for each material are shown in Table 2 (Coelho et al., 2009;Cruz et al., 2010). The data obtained were evaluated regarding the quantitative and qualitative description of the maximum principal strain for crowns, von Mises stress for screws, conical abutment and implant; and minimal principal strain for cortical and medullary bone.   Table 3. For zirconia crowns, both groups concentrated tension in the regions close to the load application point. The group M1 presented lower stress (23.878 MPa) than the group M2 (28.211 MPa) ( Figure 4A). Considering the different screws geometries used on each group, substantial differences were observed for stress, M1 exhibited 304.91 MPa concentrated between the threads and the screw head, whereas M2 exhibited considerably less tension, reaching values of 171.85 MPa in the screw head region ( Figure 4B).

Discussion
The present study revealed that conical compared to UCLA abutments present similar stress distribution for crowns, cortical and medullary bone. Conversely, the stress values were considerably higher for crowns screw and implants in the UCLA (M2) group. Thus, the null hypothesis that different abutment would not result in different stress values in implant components and surrounding bone was rejected.
In general, UCLA abutments induced higher stress in crown screws than that induced by conical abutments. These results may be related to the segmented structure, which is capable to improve biomechanical behavior with the presence of two screwed connections in the conical abutment and increase the total area for stress/strain distribution and dissipation through the components (Camargos et al., 2016). These results are consistent with previous studies, which demonstrated that non-segmented abutments generate higher stress compared to segmented abutments (Ochiai et al., 2003;Aalaei et al., 2017).
Therefore, it is important to mention the predictability of each abutment type in case of complications. Considering the UCLA abutments group, crowns are directly screwed onto implants and the higher crown screw stress found in this study may be involved in the development of complications as the treatment failure and increase in the chances of developing periimplant inflammations. For the use of UCLA abutments there are laboratorial steps, such as casting, soldering and porcelain building, or the combination of them, which can result in distortion during prosthesis manufacturing (Das Neves et al., 2014).
This is a great disadvantage because the laboratorial steps might cause a misfit. Those disadvantages can also generate biological complications such as increase of load transfer to the bone and hence bone loss, besides peri-implant infections due to bacterial accumulation in the microgap between the abutment and the implant (Dias et al., 2012). Prosthetic complications may also occur as screw loosening and fracture, which may lead to implant loss since for this type of abutment, the screw is screwed directly into the implant and its fracture may make it impossible to use the implant for rehabilitation. For this reason, it is necessary a good laboratory technician to prevent such problems from occurring. In the case of using a conical abutment, a larger contact surface area is present which allows an improved adaptation with compensation of divergences and small distortions, consequently reducing stress concentration (Ochiai et al., 2003).
Concerning the stress distribution on implants, the UCLA abutments group also demonstrated higher values compared to conical abutments. The stress concentration was mainly in the implant neck in contact with the region with the highest stress concentration of the crown screw. Previous studies reported that the design characteristics of the implant-abutment combination with resistance critical points due to the geometry along its length and alteration in cross-sectional area are responsible for higher stress and failure location (Quek et al., 2008). This may explain the higher peak of stress located at the external region of the cervical collar of the implant. In contrast, for the conical abutment group the stress on implants was approximately fourfold lower, besides this lower stress was distributed in the implant surface.
Studies suggest that the use of the prosthetic segmented abutments is directly related to a better distribution of stress (Chun et al., 2005;Aalaei et al., 2017). The implant-abutment connection is believed to play an important role in the treatment outcome. In addition, in cases of unfavorable forces segmented abutments are able to prevent fractures in the crowns screw, which is the weak component that can fracture inside the implant. Furthermore, in cases of non-segmented abutments the crown screws receive directly the occlusal loads, compromising the implant-supported prosthesis.
For zirconia crowns no differences were observed for conical and UCLA abutments. The cortical and medullary bone also did not present differences. These results are in line with a previous study that demonstrated by photoelastic stress analysis technique a similar magnitude of stresses observed for segmented and non-segmented abutment designs for single implant condition (Ochiai et al., 2003). However, other study suggest less stress and strain in the peri-implant bone in segmented abutments (Aalaei et al., 2017). These divergences may occur due to different analysis and methodologies used in studies.
Through the use of finite element analysis, we were able to numerically determine how much stress was distributed through zirconia crowns, crowns screws, conical abutment, abutment screws, implants, cortical and medullary bone in rehabilitations using conical and UCLA abutments. However, owing to the limitations of the finite element methodology, the models used in this study did not accurately represent the actual oral conditions of humans. It is necessary to consider that there are inherent limitations to in silico simulation, mainly due to assumptions regarding masticatory forces, material properties, inclination, alignment and conditions for simulating the peri-implant tissue, such as bone which is a complex dynamic structure, and its characteristics can vary substantially between individuals.

Conclusion
Within the limitations of the present study, it can be concluded that single implant-supported rehabilitations of mandibular first molars using external hexagon implants present better stress distribution on the crown screw and implants for the group of conical abutments compared to the UCLA abutment group. Further clinical studies are required to validate the results of the present study.