Medical Engineering and Physics 37 (2015) 431–445
Contents lists available at ScienceDirect
Medical Engineering and Physics journal homepage: www.elsevier.com/locate/medengphy
Number and localization of the implants for the fixed prosthetic reconstructions: On the strain in the anterior maxillary region Nilüfer Bölükbası ¸ ∗, Sinem Yeniyol Istanbul University, Faculty of Dentistry, Department of Oral Implantology, Istanbul, Turkey
a r t i c l e
i n f o
Article history: Received 21 October 2013 Revised 23 January 2015 Accepted 16 February 2015
Keywords: Finite element analysis Dental implant Anterior maxilla Biomechanics Strain
a b s t r a c t Resorption following tooth loss and poor bone quality affect the success of implants in the anterior maxilla. Inappropriate planning can cause implant loss and aesthetics problems that are difficult to resolve. There is a limited literature on the optimum number and location of implants in anterior maxilla for fabricating fixed prosthesis in biomechanical terms. This study investigated the effect of dental implant localizations in anterior maxilla on the strain values around implants using a three dimensional finite elements analysis method. Obtained strain values were compared to the data in Frost’s mechanostat theory. The entire totally edentulous maxilla was modeled using computer tomography images and five models were prepared representing different implant localizations. The distribution of implants in the models was as follows: two canines in the first model, two canines and one central incisor in the second model, two canines and central incisor in the third model, two canines and one lateral incisor in the fourth model and two canines and two lateral incisors in the fifth model. Anatomic abutments with a gingival height of 2 mm and angle of 15° were used as the abutments to fabricate one piece cemented metal fused to porcelain restoration. A chewing strength of 100 N was applied to the cingulum of all crowns at a 45° angle. Maximum strain values in all models were measured in cortical bone in implant necks. The highest strain value was measured in the first model at the cortical bone area (3037 microstrain). Except the first model, all models showed micro strain values within 1000–3000 microstrain. The fifth model was the least risky method in biomechanical terms. The results of this study should be compared with different clinical scenarios (for example different implant designs and sizes). Due to the limitations of three-dimensional finite elements analysis studies, the findings of the study need to be supported by clinical studies. © 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Implants in aesthetically important areas are defined as advanced or complex treatments [1]. Previous studies showed that the success of implants in the anterior maxilla was similar to that in other regions of the mouth [2,3]. However, there is a limited literature on the success of fixed implant supported prostheses in edentulous anterior maxilla. Furthermore, there is no consensus on the number and location of implants that should be placed in anterior maxilla. The bone in the vestibule sites of the teeth generally breaks during tooth extraction in anterior maxilla. Following the extraction, 25% is resorbed during the first year and 40–60% after 3 years [4]. Particularly due to these resorptions from vestibular to palatinal, it was necessary to localize narrow-diameter implants in more superior and palatinal locations than for natural teeth. The success of dental
∗
Corresponding author. Tel.: +90 2125323218; fax: +90 2125323254. E-mail address: [email protected], [email protected] (N. Bölükbası). ¸ http://dx.doi.org/10.1016/j.medengphy.2015.02.004 1350-4533/© 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
implants can also be affected by the spongiosis structure of the bone in the region in most cases, and by occlusal loads transmitted during movement of the mandible. Therefore, it is recommended that a minimum of three implants – two canines and one lateral or two canines and one central – should be applied on edentulous anterior maxilla [5,6]. Occlusal loads are transferred to the bone around the dental implants via the implant-supported prostheses. The loads that are transferred to the implants cause stress in the implant-bone contact area depending on the occlusal load type, size of implants, implant surface properties and structural characteristics of the bone on which the implants were applied; and implant location and prosthesis type [7]. Stress is defined as the force applied per unit area. The force that causes stress also causes strain. Strain refers to the resulting deformation or dimensional change in the bone relative to the real dimension of the bone [8]. Frost’s mechanostat hypothesis proposed that mechanical stress applied to bone cells results in the constructing of new bone or resorption [9,10]. According to Frost, at 50–1500 microstrain, remodeling is balanced; when microstrain value is 1500–3000, mild overload occurs. At this stage, any damage occurring in the bone can
432
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 1. Strain distribution in the Model 1. Implants were placed in both canine sites.
Fig. 2. Strain distribution in the Model 2. Implants were placed in both canine sites and one central incisor sites.
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
433
Fig. 3. Strain distribution in the Model 3. Implants were placed in both canine and central incisor sites.
be repaired by remodeling activity. Osseous adaptation by formation of bone can be provided. In contrast, it was reported that microstrain values of more than 3000 can cause micro cracks in the bone that cannot be repaired by remodeling. Occlusal loads that exceed the mechanical or biological loadbearing capacity of dental implants are defined as “overload” [11]. Clinical computation of the direction and magnitude of occlusal loads is difficult. In recent years, finite elements analysis (FEA) has been used to identify the loads transferred to dental implant, and the level and distribution of load in the bone around the implant. FEA analysis allows for the evaluation of various biomechanical risk factors that can affect the success of dental implants in scenarios where clinical evaluation is not possible. This study analyzed the amount and localization of deformation (strain) for implant retained fixed prosthesis with differing numbers and localizations of dental implants in anterior maxilla by means of three-dimensional finite elements analysis. The strain values
obtained in the study were compared with data for Frost’s mechanostat theory. 2. Materials and methods 2.1. Model design Computed tomography (CT) images of a totally edentulous adult patient taken from routine implant treatment planning were used to form the geometric model of the maxilla. The patient provided written consent to use the CT images in the study. The use of patient data was carried out according to the policies and procedures of the Istanbul University, Faculty of Dentistry, Department of Oral Implantology. Cone-beam Tomography (ILUMA, Orthocad, CBCT, 3M Imtec, Oklahoma, USA) was used to scan the jaw. A total of 601 crosssections were obtained at 120 kvp, 3.8 mA in 40 s. The scan was then reconstructed with volumetric data with 0.2 mm cross-section
434
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 4. Strain distribution in the Model 4. Implants were placed in both canine and one lateral incisor sites.
thickness. The cross-sections obtained after reconstruction were exported in DICOM 3.0 format. Exported cross-sections were transferred to 3D-Doctor software (Able Software Corp., MA, USA). Bone tissues on the cross-sections were separated using the “interactive segmentation” method in 3D-Doctor. Separated cross-sections were turned into a three-dimensional model using the “Complex Render” method. The obtained three-dimensional model was converted to a smooth surface using purification methods in 3D-Doctor. The resulting model consists of elements with proper ratios of modeling elements, thereby completing the modeling of the totally edentulous maxilla. The three dimensional model was exported from 3D-Doctor software in stl format. VRMesh software (VirtualGrid Inc, Bellevue City, WA, USA) was used to make dimensional and topographic arrangements on the jaw model. An arch model of the maxilla was made, based on the arch measurements conducted by Bilgin [12] on 400 Turkish patients. According to that study, the most common arch form among the Turkish population was U-shaped, medium-large and mediumlong alveolar arch. The same form was used in the present study. The alveolar crest was taken as 62.5 mm wide and 50.5 mm long. An alveolar arch of 130.68 mm was calculated. The width of the alveolar crest in the vestibular–palatinal direction was prepared as
6 mm, and the distance between the head of the crest was taken as 25 mm. Implant holes were prepared in anterior maxilla for 5 models with different implant distributions. Posterior parts of the maxilla were modeled as extended edentulous areas. The distribution of implants in the models was as follows: two canines in the first model, two canines and one central incisor in the second model, two canines and central incisor in the third model, two canines and one lateral incisor in the fourth model and two canines and two lateral incisors in the fifth model. Implant and prosthesis parts supplied in the study were scanned using a SmartOptics (Sensortechnik GmbH, Bochum, Germany) three-dimensional scanner. The models were obtained in stl format and sent to Rhinoceros 4.0 software (Seattle, WA 98103, USA). Implants (Straumann Bone Level, Switzerland) with a diameter of 4.1 mm and length of 12 mm were used for the canines and central incisors; implants with a diameter of 3.3 mm and length of 12 mm were used for lateral incisors. Anatomic abutments with a gingival height of 2 mm and angle of 15° were used as the abutments to be applied on implants. In the present study, the angle of the implants with the alveolar bone in frontal plane was planned as 10° for the canine region, 20° for the lateral incisors and 30° for the central incisors. A direct proportion formula was used to localize
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
435
Fig. 5. Strain distribution in the Model 5. Implants were placed in both canine and lateral incisor sites.
dental implants mesiodistally on the arc and to identify the distance between the implants. The alveolar arch length in the constructed jaw models was multiplied by the mean mesio-distal diameter of the crowns in natural dentition and the resulting value was divided by the mean alveolar arch length in natural dentition to calculate the mesio-distal diameter of the crowns in the constructed models. Mesio-distal diameters of the crowns in natural dentition were obtained from a study by Wheeler [13]. Mean alveolar arch length in natural dentition was obtained by the addition of mesio-distal diameters of the crowns. Based on the addition of mean mesio-distal diameters of the teeth in maxilla using the data from Wheeler, mean alveolar arch length was calculated as 128 mm. Considering mesio-distal dimensions of the teeth, the implants were located on the alveolar crest, leaving a minimum distance of 3 mm between the implants, and a minimum of 1 mm in vestibule and palatinal regions of the implants.
3. Material properties The maxilla was modeled as 1-mm cortical bone in the exterior, and as spongiosis bone in the interior according to the structural properties of type III bone [14]. In modeling of implant-supported prostheses, chromium–cobalt alloy (Wiron 99; Bego, Bremen, Germany) was used as the lower structure and feldspathic porcelain was used as the upper structure (Ceramco II; Dentsply, Burlington, ABD). Metal thickness was prepared as 0.8 mm; porcelain thickness was taken as minimum 2 mm considering crown dimensions. The modulus of elasticity and Poisson’s ratios of the materials used in this study and reference studies are presented in Table 1. 3.1. Contact management and loading The modeling performed in the Rhino program was transferred to Algor Fempro software (ALGOR, Inc., PA, USA) using three-
436
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 6. Strain distribution and values in the cortical bone in the Model 1.
Table 1 Mechanical properties of materials. Material and olacak
Elastcity modulus (GPa)
Poisson’s ratio
References
Titanium implant and abutment Cortical bone Trabecular bone (D3) Chromium–cobalt alloy Feldspathic porcelain
110 13.7 1.37 218 82.8
0.35 0.3 0.3 0.33 0.35
[15–17] [15–18] [15,18] [15,18] [15]
dimensional coordinates. The models were converted into solid model as bricks and tetrahedral elements. As many 8-node elements as possible that could be constructed in the Fempro model were used in the bricks and tetrahedral solid-modeling system. Steep and narrow
regions in the jaw models make analysis difficult, and were therefore cleaned from linear elements and thus made regular. In order to obtain realistic results, we selected the maximum number of elements permitted by the program, considering the dimensions of the jaw model. The numbers of elements and nodes that were used to model the scenarios are presented in Table 2. All models were considered as linear, homogenous and isotropic materials. Upper and lower prosthesis parts, implant screws and bone tissues were harmonized using the Boolean method in Rhino software and force transfer was provided. The model was fixed from the upper region of the jaw bone in such a way to have zero movement at each degree of freedom. In vertical loading, force was applied on all crowns from cingulum at an angle of 45°. Chewing force was taken as 100 N. The three-dimensional finite element program provided linear statistical analysis (AlgorFempro, Algor Inc., PA, USA). Strain levels
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
437
Fig. 7. Strain distribution and values in the cortical bone in the Model 2.
Table 2 The numbers of elements and nodes in 5 models.
Model 1 Model 2 Model 3 Model 4 Model 5
Elements
Nodes
380,260 908,485 501,838 925,840 524,828
69,506 163,919 91,758 166,771 94,263
were evaluated in accordance with the bone remodeling thresholds suggested by Frost’s mechanostat hypothesis. 4. Results Distributions of strain in the models following loading are presented in Figs. 1–5. Strain values in cortical and trabecular bone are presented in Figs. 6–15. It was found that strain was especially higher in the buccal region.
Maximum strain values in all models were measured in cortical bone in the neck region of the implants. Lower strain values were measured in trabecular bone than in cortical bone. Between the five models, the highest strain value was measured in the first model, in which the implants were placed in two canines. It was observed that strain values ranged between 1967 microstrain and 3037 microstrain in the cortical bone; while 714 microstrain and 2312 microstrain in the trabecular bone around the implants neck. Strain was especially concentrated at the mesiobuccal area (Figs. 1, 6, 11). When three implants are planned on the anterior maxilla rather than two, application on lateral teeth rather than central incisors will yield more advantageous results in biomechanical terms. Maximum strain values observed at the buccal site of the central incisor (1973 microstrain and 1612 microstrain at the cortical and trabecular bone respectively) in the second model whereas maximum values were located at the canine site adjacent to lateral incisor in the fourth model (Figs. 2, 4, 7, 9, 12, 14). When four implants are planned in the anterior region of the maxilla, localization of the implants in both lateral teeth regions in the canine region causes less strain than localization of the implants on both central teeth regions. On the other hand, the implants localized
438
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 8. Strain distribution and values in the cortical bone in the Model 3.
in central incisors cause higher strain values than those localized in lateral teeth region. The minimum and maximum strain values calculated in cortical and trabecular bone in the jaw were 203 and 900 microstrain respectively, located in the regions between the implants. Evaluation of the findings according Frost’s mechanostat hypothesis revealed that bone remodeling in the regions between the implants (edentulous crest) was in balance. In the first model, in which the implants are localized on both canines, microstrain values exceeded 3000 in the vestibular region in cortical bone, which suggests that resorption can occur especially in this region. In all other models, microstrain values were within the range 1000–3000 and active remodeling continued.
5. Discussion Long-term aesthetic and functional success in rehabilitation of the maxilla with dental implants depend on a detailed treatment planning. In this study, the effect of implant localizations on implant health in anterior maxilla was analyzed using FEA. A review of the literature shows a limited research on FEA of the maxilla compared with studies of the mandible. This is mainly because the maxilla cannot be realistically modeled, due to its complex anatomic structure. Therefore, previous three-dimensional FEA analyses of the maxilla used segmental models with few elements and nodes in order to simplify the models [19,20]. Okumura et al. [21] modeled the entire maxilla with conventional segmental models and compared strain distributions. They found that
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
439
Fig. 9. Strain distribution and values in the cortical bone in the Model 4.
strain distribution was similar with FEA analysis that modeled the entire maxilla in segmental models. However, studies that reported significant strain values and strain distribution also revealed that it would be appropriate to model the entire maxilla. In the present study, we modeled the entire maxilla to obtain realistic data that can be applied to clinic. The aim of our study was to provide biomechanical reference to clinicians in planning implant numbers and localizations in the anterior maxilla for the treatment of totally and partially edentulous patients. Prosthetic planning is generally achieved in three segmental prosthetic planning (anterior maxilla and the bilateral posterior maxilla segments) for the full edentulous individuals. There is adequate clinical data in terms of implant number and localization for esthetic demands in the anterior edentulous maxillary segment, whereas no biomechanical evidential data are present for biomechanical construction. In cases of maxillary full or partial edentulism, various treatment options can be planned according to the implant numbers and localizations. It is almost impossible to evaluate each of these biomechanical analyses and publish them in a whole in one manuscript. For this reason, we chose to plan a biomechanical analysis for the anterior edentulous maxillary segment since there is no evident data about the implant number and localizations specific to this segment of the
maxillary arch. In our study, the implants were placed in the anterior maxilla and posterior sites were modeled as edentulous areas. In this regard, our study should be further evaluated to contribute to a better understanding for the biomechanical performance of bilateral edentulous maxillary segments clinically. In addition, this type of modeling reduced the element and node numbers, and, simplified the model for an effective analysis. Recent studies using FEA commonly use CT images to model the dimensions and forms of the jaw [20,22–26]. CT images with 0.2-mm cross-section, obtained from an individual with edentulous maxilla, were used to precisely represent the anatomy of the maxilla. Studies using tomography images risk producing erroneous results if the models reproduce deformations in the jaw of the patient. It was believed that performing the study on an appropriate ideal jaw model rather than the jaw model of a specific person would yield results that were more appropriate for clinical use. Therefore, a new upper jaw model was prepared based on the tomography image. As the present study used specific dimensions and a U-shaped alveolar arch, which is the most common shape among the Turkish population, the results can be compared with prospective data from other alveolar arch forms in future studies. Sagat et al. [27] reported that, because the distance between implants differs for prostheses implanted in arches with different dimensions and forms, the strain
440
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 10. Strain distribution and values in the cortical bone in the Model 5.
values measured on peri-implant bone can vary. The researchers used FE strain analysis to determine the most advantageous implant localizations in fixed-implant-supported prostheses that were supported by 6 or 8 implants in edentulous maxilla with different alveolar arch forms. Among all alveolar arch models, the most favorable strain values were obtained for implant groups on canines and lateral incisors. Comparison of the present findings with values from Frost’s mechanostat theory showed that strain values for edentulous areas that were not supported by an implant (edentulous areas) were within physiologic limits. This study evaluated 5 implant localizations that are commonly used in clinic. Apart from these models involving implant localizations on incisors and canines (a total of 6 implants), other scenarios can be modeled. It is presumed that the strain value per implant and the implant value will decrease. However, considering the
distance between implants and the resorption pattern in the anterior maxilla, this type of a planning can give rise to aesthetic problems. Sano et al. [24] evaluated the number of implants and the stress on the bone caused by localizations in an edentulous maxilla, and reported that the level of stress was determined by whether the implants were connected to each other. In that study, dental implants were placed in the locations as 14 unsplinted implants (S14), 6 splinted implants (canine, premolar, and molar regions S6), 4 splinted implants (S4), and 6 anterior implants (incisors and canines, A6). The S6 model showed similar levels of stress and deformation to the US 14 and S14 models. Resorption following the loss of natural teeth makes it difficult to achieve ideal implant applications in the anterior maxilla. Dental implants must be located on more palatinal and superior positions than natural teeth. Sadrimanesh et al. [26] analyzed stress distribution
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
441
Fig. 11. Strain distribution and values in the trabecular bone in the Model 1.
around implants in the anterior maxilla with different labial angles; they reported that as fixture angle increased, stress value measured on cortical bone in the labial of implants increased. In our study, considering resorption following tooth loss, the angle of the implants to frontal places was planned as 10° in canines, 20° in lateral incisors and 30° in central incisors. Taking into account the localization angles of the implants, abutments with 15° angle were preferred. Tian et al. [28] conducted a three-dimensional FEA and reported that the use of angled abutments decreased stress when the implants could not be localized in ideal positions. It should be remembered that the abutment angles used in different clinical scenarios—including implants localized at ideal positions or different positions—will affect the level of stress imposed on the peri-implant bone. There is a limited clinical and radiological study in the literature on the localization of dental implants in vestibule–palatinal and
mesial–distal directions. It is considered that the distance between the implants should be minimum 3 mm in order to prevent potential resorption in bone [29]. Tarnow et al. [29] reported that when the distance between two implants exceeds 3 mm, crestal bone loss is 0.45 mm; when it is less than 3 mm, crestal bone loss is 1.04 mm. It is suggested that a minimum of 1 mm of bone should be left in the vestibule and palatinal surfaces of implants in order to ensure an appropriate exit profile of the implant from the alveolar crest; and for sufficient feeding of the bone. Spray et al. [30] reported that resorptions decreased and even bone apposition was observed when there is 1.8 mm or more of bone in the vestibule surfaces of implants. Bidez and Misch reported that strain in bone was lower in cases with 3 implants when compared to those with 2 implants [31,32]. Considering the above factors and clinical practices, the present study modeled 5 different implant localizations leaving a minimum dis-
442
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 12. Strain distribution and values in the trabecular bone in the Model 2.
tance of 3 mm between the implants and a minimum of 1 mm in vestibule and palatinal regions of implants. There is a large body of research analyzing the amount of stress transferred to dental implants and peri-implant bone by the materials used in implant-supported prostheses [33–37]. Considering previous research and clinical practices, the present study chromium– cobalt alloy (Wiron 99; Bego, Bremen, Germany) [33,38,39] supported feldspastic porcelain (Ceramco II; Dentsply, Burlington, ABD) [33,34,40] were used as implant supported prosthesis. FE studies in oral implantology applied various occlusal forces of 100–2000 N [18,22,27,41–43]. In our study, 100 N mean vertical chewing force was used [42]. There are a limited number of previous studies on loading in different regions of the anterior maxilla. Clelland et al. [22] applied occlusal loads in the direction of the longer axis of implants. Saab et al. [43] applied occlusal loads from the cingulum region, and Hsu et al. [44] applied the occlusal loads at angles of 0°, 30° and 60° to the longer axis of the implant. Forces applied along the longer axis of implants represent the situation when the upper and lower incisors are closed. This type of closing during chewing
is observed in a small proportion of patients with distorted upper– lower jaw relationship. Furthermore, the measured chewing force is lower when the mandible is in eccentric position and the posterior teeth do not contact [43]. The application of vertical loads in buccal and apical direction from the cingulum regions of crowns reflects the location of mandibular incisors closest to the palatinal surfaces of maxillary incisors; in other words, centric occlusion location. Centric occlusion position is the appropriate area in which to make examine the magnitude of vertical loading applied to the anterior maxilla [43]. Therefore, in the present study, vertical chewing forces were applied at a 45° angle from cingulum regions. Force application over dental implants, abutment or crowns affects the reliability of the results. Application of chewing forces to the crowns provides realistic results [44]. For this reason, the present study applied loads to the crowns. It is difficult to make calculations according to permanent deformations in order to analyze an element. Instead, the analysis treats the systems as linear elastic; in other words, it is more convenient to analyze an idealized model of the system. In the present
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
443
Fig. 13. Strain distribution and values in the trabecular bone in the Model 3.
study, the upper jaw was considered as linear elastic for all calculations. Furthermore, the maxilla was considered as homogenous and isotropic, but is not a homogenous structure in practice. Similarly, there are varying opinions on whether the jaws are isotropic or orthotropic [23,45,46]. In a study that modeled the jaw as orthotropic, stress increased by 25%. In this study, osseointegration was assumed to be 100% in order to reduce calculation time. However, osseointegration and stress in peri-implant bones is proportional [47]. Spivey et al. [47] reported that when osseointegration was 83.3% strain measured in peri-implant bone increased by 5% compared to the situation when osseointegration was considered as 100%. One limitation of our study is the fact that Frost’s mechanostat theory was not specifically developed for dental implants. Comprehensive researches are still needed to determine bone strain values
under occlusal loads for peri-implant bone, although Frost himself notified that human load-bearing bones adapt to mechanical changes [7–9]. The most important disadvantage of FEA analyses is that the results are only valid for the selected model. For example, assumptions are made for variable data such as the dimensions of dental implants, cortical bone thickness, trabecular bone density and amount of osseointegration. Therefore, it should be remembered that the results will vary when the input data change. 6. Conclusion This study evaluated biomechanical behavior for implant localizations of fabricating implant retained fixed prostheses, in the anterior maxilla. Our findings reveal that increasing the number of implants
444
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 14. Strain distribution and values in the trabecular bone in the Model 4.
Fig. 15. Strain distribution and values in the trabecular bone in the Model 5.
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
reduced the risk of bone resorption. Different model types should be evaluated and prospective clinical studies with long-term follow-up are required for definitive conclusions. Conflict of interests None declared. Funding None. Ethical approval The use of patient data was carried out according to the policies and procedures of the Istanbul University, Faculty of Dentistry, Department of Oral Implantology. The patient gave written consent to use the CT images for the purposes of this study. References [1] Buser D, Martin W, Belser UC. Optimizing esthetics for implant restorations in the anterio maxilla: anatomic and surgical considerations. Int J Oral Maxillofac Implants 2004;19(suppl):43–61. [2] Belser UC, Buser D, Higginbottom F. Consensus statements and recommended clinical procedures regarding esthetics in implant dentistry. Int J Oral Maxillofac Implants 2004;19(suppl):73–4. [3] Belser UC, Schmid B, Higginbottom F, Buser D. Outcome analysis of implant restorations located in the anterior maxilla: a review of the recent literature. Int J Oral Maxillofac Implants 2004;19(suppl):30–42. [4] Pietrokowski J. The bony residual ridge in man. J Prosthet Dent 1975;34:456–62. [5] Misch CE, Bidez MW. Implant protected occlusion a biomechanical rationale. Compend Cont Educ Dent 1994;15:1330–43. [6] Misch CE. Premaxilla implant considerations: surgery and fixed prosthodontics. In: Misch CE, editor. Contemporary implant dentistry. St Louis: Mosby; 1993. p. 575–97. [7] Stanford CM, Brand RA. Toward an understanding of implant occlusion and strain adaptive bone modelling and remodelling. J Prosthet Dent 1999;81:553–61. [8] Fung YC. Biomechanics: Mechanical properties of living tissues. 3rd ed. New York: Springer- Verlag; 1993. p. 1–58. [9] Frost HM. Perspectives: bone’s mechanical usage windows. Bone Miner 1992;19:257–71. [10] Frost HM. A 2003 update of bone physiolology and Wollf’s law for clinicians. Angle Orthod 2004;74:3–15. [11] Isidor F. Occlusal loading in implant dentistry. In: Lang NP, Karring T, Lindhe J, editors. Proceedings of the third European workshop on periodontology—implant dentistry. Berlin: Quientessence Books; 1999. p. 358–75. [12] Bilgin T. Research concerning about the optimum prefabricated impression trays for total edentulous patients in Turkey [Ph.D. thesis]. Istanbul, Turkey: Istanbul University; 1989. [13] Ash MM. Wheeler’s dental anatomy, physiology and occlusion. Philadelphia, PA: WB Saunders; 1993. p. 128–273. [14] Lekholm U, Zarb GA. Patient selection and preparation. In: Branemark P-I, Zarb GA, Albrektsson T, editors. Tissue integrated prostheses: osseointegration in clinical dentistry. Chicago, IL: Quintessence; 1985. p. 199–209. [15] Sevimay M, Turhan F, Kılıçaslan MA, Eskitasçıoglu G. Three-dimensional finite element analysis of the different bone quality on stress distribution in an implantsupported crown. J Prosthet Dent 2005;93(3):227–34. [16] Himmlova L, Dostolava T, Kacovsky A, Konvickova S. Influence of implant length and diameter on stress distribution: a finite element analysis. J Prosthet Dent 2004;91(1):20–5. [17] I˙ plikçiolu H, Akça K. Comparative evaluation of the effect of diameter, length and number of implants supporting three-unit fixed partial prostheses on stress distribution in the bone. J Dent 2002;30(1):41–6. [18] Çaglar A, Aydın C, Yılmaz C, Korkmaz T. Effects of mesio distal inclination of implants on stress distribution in implant-supported fixed prostheses. Int J Oral Maxillofac Implants 2006;21:36–44. [19] Tepper G, Haas R, Zechner W, Krach W, Watzek G. Three-dimensional finite element analysis of implant stability in the atrophic posterior maxilla: a mathematical study of the sinus floor augmentation. Clin Oral Implants Res 2002;13:657–65.
445
[20] Lin CL, Wang JC, Ramp LC, Liu PR. Biomechanical response of implant systems placed in the maxillary posterior region under various conditions of angulation, bone density, and loading. Int J Oral Maxillofac Implants 2008;23:57–64. [21] Okumura N, Stegaroiu R, Nishiyama H, Kurokawa K, Kitamura E, Hayashi T, et al. Finite element analysis of implant-embedded maxilla model from CT data: comparison with the conventional model. J Prosthodont Res 2011;55:24–31. [22] Clelland NL, Lee KL, Bimbenet OC, Brantley WA. A three-dimensional finite element stress analysis of angled abutments for an implant placed in the anterior maxilla. J Prosthodont 1995;4:95–100. [23] Clelland NL, Lee JK, Bimbenet OC, Gilat A. Use of an axisymmetric finite element method to compare maxillary bone variables for a loaded implant. J Prosthodont 1993;2:183–9. [24] Sano M, Ikebe K, Yang TC, Maeda Y. Biomechanical rationale for six splinted implants in bilateral canine, premolar, and molar regions in an edentulous maxilla. Implant Dent 2012;21:220–4. [25] Lee JS, Lim YJ. Three-dimensional numerical simulation of stress induced by different lengths of osseointegrated implants in the anterior maxilla. Comput Methods Biomech Biomed Eng 2013;16(11):1143–9. [26] Sadrimanesh R, Siadat H, Sadr-Eshkevari P, Monzavi A, Maurer P, Rashad A. Alveolar bone stress around implants with different abutment angulation: an FEanalysis of anterior maxilla. Implant Dent 2012;21:196–201. [27] Sagat G, Yalcin S, Gultekin BA, Mijiritsky E. Influence of arch shape and implant position on stress distribution around implants supporting fixed full-arch prosthesis in edentulous maxilla. Implant Dent 2010;19:498–508. [28] Tian K, Chen J, Han L, Yang J, Huang W, Wu D. Angled abutments result in increased or decreased stress on surrounding bone of single-unit dental implants: a finite element analysis. Med Eng Phys 2012;34:1526–31. [29] Tarnow DP, Cho SC, Wallace SS. The effect of inter-implant distance on the height of inter-implant bone crest. J Periodontol 2000;71:546–9. [30] Spray JR, Black CG, Morris HF, Ochi S. The influence of bone thickness on facial marginal bone response: stage 1 placement through stage 2 uncovering. Ann Periodontol 2000;5:119–28. [31] Bidez MW, Misch CE. The biomechanics of inter-implant spacing; 1990. [32] Bidez MW, Misch CE. Issues in bone mechanics related to oral implants. Implant Dent 1992;1:289–94. [33] Sevimay M, Üsümez A, Eskitasçıoglu G. The influence of various occlusal materials on stresses transferred to implant-supported prostheses and supporting bone: a three-dimensional finite-element study. J Biomed Mater Res Part B Appl Biomater 2005;73:140–7. [34] Cibirka RM, Razzoog ME, Lang BR, Stohler CS. Determining the force absorption quotient for restorative materials used in implant occlusal surfaces. J Prosthet Dent 1992;67:361–4. [35] Eskitascıoglu G, Baran I, Aykac Y, Oztas D. Investigation of the effect of different esthetic materials in implant-crown design. Turkish J Oral Imp 1996;4:13–19. [36] Sertgöz A. Finite element analysis study of the effect of superstructure material on stress distribution in implant supported fixed prosthesis. Int J Prosthodont 1997;10:19–27. [37] Papavasiliou G, Kamposiora P, Bayne SC, Felton DA. Three dimensional finite element analysis of stress distribution around single tooth implants as a function of bony support, prosthesis type and loading during function. J Prosthet Dent 1996;76:633–40. [38] Wi1lliams KR, Watson CJ, Murphy WM, Scott J, Gregory M, Sinobad D. Finite element analysis of fixed prostheses attached to osseointegrated implants. Quintessence Int 1990;21:563–70. [39] O’Brien WJ. Dental materials and their selection. 2nd ed. Chicago: Quintessence; 1997. p. 259–72. [40] Hojjatie B, Anusavice KJ. Three-dimensional finite element analysis of glassceramic dental crowns. J Biomech 1990;23:1157–66. [41] Bozkaya D, Müftü S, Müftü A. Evaluation of load transfer characteristics of five different implants in compact bone at different load levels by finite elements analysis. J Prosthet Dent 2004;9:523–30. [42] Tada S, Stegaroiu R, Kitamura E, Miyakawa O, Kusakari H. Influence of implant design and bone quality on stress/strain distribution in bone around implants: a 3dimensional finite element analysis. Int J Oral Maxillofac Implants 2003;18:357– 68. [43] Saab XE, Griggs JA, Powers JM, Engelmeier RL. Effect of abutment angulation on the strain on the bone around an implant in the anterior maxilla: a finite element study. J Prosthet Dent 2007;97:85–92. [44] Hsu ML, Chen FC, Kao HC, Cheng CK. Influence of off-axis loading of an anterior maxillary implant: a three-dimensional finite element analysis. Int J Oral Maxillofac Implants 2007;22:301–9. [45] Ashman RB, Van Buskirk WC. The elastic properties of a human mandible. Adv Dent Res 1987;1:64–7. [46] Cowin SC. Bone mechanics. Boca Raton, FL: CRC; 1989. p. 27–141. [47] Spivey JD, Kong W, Fotos PG. Stress distribution at the bone-to-implant interface. a 3-D finite element analyses (Abstract 106). J Dent Res 1993;72:117.
Contents lists available at ScienceDirect
Medical Engineering and Physics journal homepage: www.elsevier.com/locate/medengphy
Number and localization of the implants for the fixed prosthetic reconstructions: On the strain in the anterior maxillary region Nilüfer Bölükbası ¸ ∗, Sinem Yeniyol Istanbul University, Faculty of Dentistry, Department of Oral Implantology, Istanbul, Turkey
a r t i c l e
i n f o
Article history: Received 21 October 2013 Revised 23 January 2015 Accepted 16 February 2015
Keywords: Finite element analysis Dental implant Anterior maxilla Biomechanics Strain
a b s t r a c t Resorption following tooth loss and poor bone quality affect the success of implants in the anterior maxilla. Inappropriate planning can cause implant loss and aesthetics problems that are difficult to resolve. There is a limited literature on the optimum number and location of implants in anterior maxilla for fabricating fixed prosthesis in biomechanical terms. This study investigated the effect of dental implant localizations in anterior maxilla on the strain values around implants using a three dimensional finite elements analysis method. Obtained strain values were compared to the data in Frost’s mechanostat theory. The entire totally edentulous maxilla was modeled using computer tomography images and five models were prepared representing different implant localizations. The distribution of implants in the models was as follows: two canines in the first model, two canines and one central incisor in the second model, two canines and central incisor in the third model, two canines and one lateral incisor in the fourth model and two canines and two lateral incisors in the fifth model. Anatomic abutments with a gingival height of 2 mm and angle of 15° were used as the abutments to fabricate one piece cemented metal fused to porcelain restoration. A chewing strength of 100 N was applied to the cingulum of all crowns at a 45° angle. Maximum strain values in all models were measured in cortical bone in implant necks. The highest strain value was measured in the first model at the cortical bone area (3037 microstrain). Except the first model, all models showed micro strain values within 1000–3000 microstrain. The fifth model was the least risky method in biomechanical terms. The results of this study should be compared with different clinical scenarios (for example different implant designs and sizes). Due to the limitations of three-dimensional finite elements analysis studies, the findings of the study need to be supported by clinical studies. © 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Implants in aesthetically important areas are defined as advanced or complex treatments [1]. Previous studies showed that the success of implants in the anterior maxilla was similar to that in other regions of the mouth [2,3]. However, there is a limited literature on the success of fixed implant supported prostheses in edentulous anterior maxilla. Furthermore, there is no consensus on the number and location of implants that should be placed in anterior maxilla. The bone in the vestibule sites of the teeth generally breaks during tooth extraction in anterior maxilla. Following the extraction, 25% is resorbed during the first year and 40–60% after 3 years [4]. Particularly due to these resorptions from vestibular to palatinal, it was necessary to localize narrow-diameter implants in more superior and palatinal locations than for natural teeth. The success of dental
∗
Corresponding author. Tel.: +90 2125323218; fax: +90 2125323254. E-mail address: [email protected], [email protected] (N. Bölükbası). ¸ http://dx.doi.org/10.1016/j.medengphy.2015.02.004 1350-4533/© 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
implants can also be affected by the spongiosis structure of the bone in the region in most cases, and by occlusal loads transmitted during movement of the mandible. Therefore, it is recommended that a minimum of three implants – two canines and one lateral or two canines and one central – should be applied on edentulous anterior maxilla [5,6]. Occlusal loads are transferred to the bone around the dental implants via the implant-supported prostheses. The loads that are transferred to the implants cause stress in the implant-bone contact area depending on the occlusal load type, size of implants, implant surface properties and structural characteristics of the bone on which the implants were applied; and implant location and prosthesis type [7]. Stress is defined as the force applied per unit area. The force that causes stress also causes strain. Strain refers to the resulting deformation or dimensional change in the bone relative to the real dimension of the bone [8]. Frost’s mechanostat hypothesis proposed that mechanical stress applied to bone cells results in the constructing of new bone or resorption [9,10]. According to Frost, at 50–1500 microstrain, remodeling is balanced; when microstrain value is 1500–3000, mild overload occurs. At this stage, any damage occurring in the bone can
432
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 1. Strain distribution in the Model 1. Implants were placed in both canine sites.
Fig. 2. Strain distribution in the Model 2. Implants were placed in both canine sites and one central incisor sites.
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
433
Fig. 3. Strain distribution in the Model 3. Implants were placed in both canine and central incisor sites.
be repaired by remodeling activity. Osseous adaptation by formation of bone can be provided. In contrast, it was reported that microstrain values of more than 3000 can cause micro cracks in the bone that cannot be repaired by remodeling. Occlusal loads that exceed the mechanical or biological loadbearing capacity of dental implants are defined as “overload” [11]. Clinical computation of the direction and magnitude of occlusal loads is difficult. In recent years, finite elements analysis (FEA) has been used to identify the loads transferred to dental implant, and the level and distribution of load in the bone around the implant. FEA analysis allows for the evaluation of various biomechanical risk factors that can affect the success of dental implants in scenarios where clinical evaluation is not possible. This study analyzed the amount and localization of deformation (strain) for implant retained fixed prosthesis with differing numbers and localizations of dental implants in anterior maxilla by means of three-dimensional finite elements analysis. The strain values
obtained in the study were compared with data for Frost’s mechanostat theory. 2. Materials and methods 2.1. Model design Computed tomography (CT) images of a totally edentulous adult patient taken from routine implant treatment planning were used to form the geometric model of the maxilla. The patient provided written consent to use the CT images in the study. The use of patient data was carried out according to the policies and procedures of the Istanbul University, Faculty of Dentistry, Department of Oral Implantology. Cone-beam Tomography (ILUMA, Orthocad, CBCT, 3M Imtec, Oklahoma, USA) was used to scan the jaw. A total of 601 crosssections were obtained at 120 kvp, 3.8 mA in 40 s. The scan was then reconstructed with volumetric data with 0.2 mm cross-section
434
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 4. Strain distribution in the Model 4. Implants were placed in both canine and one lateral incisor sites.
thickness. The cross-sections obtained after reconstruction were exported in DICOM 3.0 format. Exported cross-sections were transferred to 3D-Doctor software (Able Software Corp., MA, USA). Bone tissues on the cross-sections were separated using the “interactive segmentation” method in 3D-Doctor. Separated cross-sections were turned into a three-dimensional model using the “Complex Render” method. The obtained three-dimensional model was converted to a smooth surface using purification methods in 3D-Doctor. The resulting model consists of elements with proper ratios of modeling elements, thereby completing the modeling of the totally edentulous maxilla. The three dimensional model was exported from 3D-Doctor software in stl format. VRMesh software (VirtualGrid Inc, Bellevue City, WA, USA) was used to make dimensional and topographic arrangements on the jaw model. An arch model of the maxilla was made, based on the arch measurements conducted by Bilgin [12] on 400 Turkish patients. According to that study, the most common arch form among the Turkish population was U-shaped, medium-large and mediumlong alveolar arch. The same form was used in the present study. The alveolar crest was taken as 62.5 mm wide and 50.5 mm long. An alveolar arch of 130.68 mm was calculated. The width of the alveolar crest in the vestibular–palatinal direction was prepared as
6 mm, and the distance between the head of the crest was taken as 25 mm. Implant holes were prepared in anterior maxilla for 5 models with different implant distributions. Posterior parts of the maxilla were modeled as extended edentulous areas. The distribution of implants in the models was as follows: two canines in the first model, two canines and one central incisor in the second model, two canines and central incisor in the third model, two canines and one lateral incisor in the fourth model and two canines and two lateral incisors in the fifth model. Implant and prosthesis parts supplied in the study were scanned using a SmartOptics (Sensortechnik GmbH, Bochum, Germany) three-dimensional scanner. The models were obtained in stl format and sent to Rhinoceros 4.0 software (Seattle, WA 98103, USA). Implants (Straumann Bone Level, Switzerland) with a diameter of 4.1 mm and length of 12 mm were used for the canines and central incisors; implants with a diameter of 3.3 mm and length of 12 mm were used for lateral incisors. Anatomic abutments with a gingival height of 2 mm and angle of 15° were used as the abutments to be applied on implants. In the present study, the angle of the implants with the alveolar bone in frontal plane was planned as 10° for the canine region, 20° for the lateral incisors and 30° for the central incisors. A direct proportion formula was used to localize
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
435
Fig. 5. Strain distribution in the Model 5. Implants were placed in both canine and lateral incisor sites.
dental implants mesiodistally on the arc and to identify the distance between the implants. The alveolar arch length in the constructed jaw models was multiplied by the mean mesio-distal diameter of the crowns in natural dentition and the resulting value was divided by the mean alveolar arch length in natural dentition to calculate the mesio-distal diameter of the crowns in the constructed models. Mesio-distal diameters of the crowns in natural dentition were obtained from a study by Wheeler [13]. Mean alveolar arch length in natural dentition was obtained by the addition of mesio-distal diameters of the crowns. Based on the addition of mean mesio-distal diameters of the teeth in maxilla using the data from Wheeler, mean alveolar arch length was calculated as 128 mm. Considering mesio-distal dimensions of the teeth, the implants were located on the alveolar crest, leaving a minimum distance of 3 mm between the implants, and a minimum of 1 mm in vestibule and palatinal regions of the implants.
3. Material properties The maxilla was modeled as 1-mm cortical bone in the exterior, and as spongiosis bone in the interior according to the structural properties of type III bone [14]. In modeling of implant-supported prostheses, chromium–cobalt alloy (Wiron 99; Bego, Bremen, Germany) was used as the lower structure and feldspathic porcelain was used as the upper structure (Ceramco II; Dentsply, Burlington, ABD). Metal thickness was prepared as 0.8 mm; porcelain thickness was taken as minimum 2 mm considering crown dimensions. The modulus of elasticity and Poisson’s ratios of the materials used in this study and reference studies are presented in Table 1. 3.1. Contact management and loading The modeling performed in the Rhino program was transferred to Algor Fempro software (ALGOR, Inc., PA, USA) using three-
436
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 6. Strain distribution and values in the cortical bone in the Model 1.
Table 1 Mechanical properties of materials. Material and olacak
Elastcity modulus (GPa)
Poisson’s ratio
References
Titanium implant and abutment Cortical bone Trabecular bone (D3) Chromium–cobalt alloy Feldspathic porcelain
110 13.7 1.37 218 82.8
0.35 0.3 0.3 0.33 0.35
[15–17] [15–18] [15,18] [15,18] [15]
dimensional coordinates. The models were converted into solid model as bricks and tetrahedral elements. As many 8-node elements as possible that could be constructed in the Fempro model were used in the bricks and tetrahedral solid-modeling system. Steep and narrow
regions in the jaw models make analysis difficult, and were therefore cleaned from linear elements and thus made regular. In order to obtain realistic results, we selected the maximum number of elements permitted by the program, considering the dimensions of the jaw model. The numbers of elements and nodes that were used to model the scenarios are presented in Table 2. All models were considered as linear, homogenous and isotropic materials. Upper and lower prosthesis parts, implant screws and bone tissues were harmonized using the Boolean method in Rhino software and force transfer was provided. The model was fixed from the upper region of the jaw bone in such a way to have zero movement at each degree of freedom. In vertical loading, force was applied on all crowns from cingulum at an angle of 45°. Chewing force was taken as 100 N. The three-dimensional finite element program provided linear statistical analysis (AlgorFempro, Algor Inc., PA, USA). Strain levels
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
437
Fig. 7. Strain distribution and values in the cortical bone in the Model 2.
Table 2 The numbers of elements and nodes in 5 models.
Model 1 Model 2 Model 3 Model 4 Model 5
Elements
Nodes
380,260 908,485 501,838 925,840 524,828
69,506 163,919 91,758 166,771 94,263
were evaluated in accordance with the bone remodeling thresholds suggested by Frost’s mechanostat hypothesis. 4. Results Distributions of strain in the models following loading are presented in Figs. 1–5. Strain values in cortical and trabecular bone are presented in Figs. 6–15. It was found that strain was especially higher in the buccal region.
Maximum strain values in all models were measured in cortical bone in the neck region of the implants. Lower strain values were measured in trabecular bone than in cortical bone. Between the five models, the highest strain value was measured in the first model, in which the implants were placed in two canines. It was observed that strain values ranged between 1967 microstrain and 3037 microstrain in the cortical bone; while 714 microstrain and 2312 microstrain in the trabecular bone around the implants neck. Strain was especially concentrated at the mesiobuccal area (Figs. 1, 6, 11). When three implants are planned on the anterior maxilla rather than two, application on lateral teeth rather than central incisors will yield more advantageous results in biomechanical terms. Maximum strain values observed at the buccal site of the central incisor (1973 microstrain and 1612 microstrain at the cortical and trabecular bone respectively) in the second model whereas maximum values were located at the canine site adjacent to lateral incisor in the fourth model (Figs. 2, 4, 7, 9, 12, 14). When four implants are planned in the anterior region of the maxilla, localization of the implants in both lateral teeth regions in the canine region causes less strain than localization of the implants on both central teeth regions. On the other hand, the implants localized
438
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 8. Strain distribution and values in the cortical bone in the Model 3.
in central incisors cause higher strain values than those localized in lateral teeth region. The minimum and maximum strain values calculated in cortical and trabecular bone in the jaw were 203 and 900 microstrain respectively, located in the regions between the implants. Evaluation of the findings according Frost’s mechanostat hypothesis revealed that bone remodeling in the regions between the implants (edentulous crest) was in balance. In the first model, in which the implants are localized on both canines, microstrain values exceeded 3000 in the vestibular region in cortical bone, which suggests that resorption can occur especially in this region. In all other models, microstrain values were within the range 1000–3000 and active remodeling continued.
5. Discussion Long-term aesthetic and functional success in rehabilitation of the maxilla with dental implants depend on a detailed treatment planning. In this study, the effect of implant localizations on implant health in anterior maxilla was analyzed using FEA. A review of the literature shows a limited research on FEA of the maxilla compared with studies of the mandible. This is mainly because the maxilla cannot be realistically modeled, due to its complex anatomic structure. Therefore, previous three-dimensional FEA analyses of the maxilla used segmental models with few elements and nodes in order to simplify the models [19,20]. Okumura et al. [21] modeled the entire maxilla with conventional segmental models and compared strain distributions. They found that
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
439
Fig. 9. Strain distribution and values in the cortical bone in the Model 4.
strain distribution was similar with FEA analysis that modeled the entire maxilla in segmental models. However, studies that reported significant strain values and strain distribution also revealed that it would be appropriate to model the entire maxilla. In the present study, we modeled the entire maxilla to obtain realistic data that can be applied to clinic. The aim of our study was to provide biomechanical reference to clinicians in planning implant numbers and localizations in the anterior maxilla for the treatment of totally and partially edentulous patients. Prosthetic planning is generally achieved in three segmental prosthetic planning (anterior maxilla and the bilateral posterior maxilla segments) for the full edentulous individuals. There is adequate clinical data in terms of implant number and localization for esthetic demands in the anterior edentulous maxillary segment, whereas no biomechanical evidential data are present for biomechanical construction. In cases of maxillary full or partial edentulism, various treatment options can be planned according to the implant numbers and localizations. It is almost impossible to evaluate each of these biomechanical analyses and publish them in a whole in one manuscript. For this reason, we chose to plan a biomechanical analysis for the anterior edentulous maxillary segment since there is no evident data about the implant number and localizations specific to this segment of the
maxillary arch. In our study, the implants were placed in the anterior maxilla and posterior sites were modeled as edentulous areas. In this regard, our study should be further evaluated to contribute to a better understanding for the biomechanical performance of bilateral edentulous maxillary segments clinically. In addition, this type of modeling reduced the element and node numbers, and, simplified the model for an effective analysis. Recent studies using FEA commonly use CT images to model the dimensions and forms of the jaw [20,22–26]. CT images with 0.2-mm cross-section, obtained from an individual with edentulous maxilla, were used to precisely represent the anatomy of the maxilla. Studies using tomography images risk producing erroneous results if the models reproduce deformations in the jaw of the patient. It was believed that performing the study on an appropriate ideal jaw model rather than the jaw model of a specific person would yield results that were more appropriate for clinical use. Therefore, a new upper jaw model was prepared based on the tomography image. As the present study used specific dimensions and a U-shaped alveolar arch, which is the most common shape among the Turkish population, the results can be compared with prospective data from other alveolar arch forms in future studies. Sagat et al. [27] reported that, because the distance between implants differs for prostheses implanted in arches with different dimensions and forms, the strain
440
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 10. Strain distribution and values in the cortical bone in the Model 5.
values measured on peri-implant bone can vary. The researchers used FE strain analysis to determine the most advantageous implant localizations in fixed-implant-supported prostheses that were supported by 6 or 8 implants in edentulous maxilla with different alveolar arch forms. Among all alveolar arch models, the most favorable strain values were obtained for implant groups on canines and lateral incisors. Comparison of the present findings with values from Frost’s mechanostat theory showed that strain values for edentulous areas that were not supported by an implant (edentulous areas) were within physiologic limits. This study evaluated 5 implant localizations that are commonly used in clinic. Apart from these models involving implant localizations on incisors and canines (a total of 6 implants), other scenarios can be modeled. It is presumed that the strain value per implant and the implant value will decrease. However, considering the
distance between implants and the resorption pattern in the anterior maxilla, this type of a planning can give rise to aesthetic problems. Sano et al. [24] evaluated the number of implants and the stress on the bone caused by localizations in an edentulous maxilla, and reported that the level of stress was determined by whether the implants were connected to each other. In that study, dental implants were placed in the locations as 14 unsplinted implants (S14), 6 splinted implants (canine, premolar, and molar regions S6), 4 splinted implants (S4), and 6 anterior implants (incisors and canines, A6). The S6 model showed similar levels of stress and deformation to the US 14 and S14 models. Resorption following the loss of natural teeth makes it difficult to achieve ideal implant applications in the anterior maxilla. Dental implants must be located on more palatinal and superior positions than natural teeth. Sadrimanesh et al. [26] analyzed stress distribution
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
441
Fig. 11. Strain distribution and values in the trabecular bone in the Model 1.
around implants in the anterior maxilla with different labial angles; they reported that as fixture angle increased, stress value measured on cortical bone in the labial of implants increased. In our study, considering resorption following tooth loss, the angle of the implants to frontal places was planned as 10° in canines, 20° in lateral incisors and 30° in central incisors. Taking into account the localization angles of the implants, abutments with 15° angle were preferred. Tian et al. [28] conducted a three-dimensional FEA and reported that the use of angled abutments decreased stress when the implants could not be localized in ideal positions. It should be remembered that the abutment angles used in different clinical scenarios—including implants localized at ideal positions or different positions—will affect the level of stress imposed on the peri-implant bone. There is a limited clinical and radiological study in the literature on the localization of dental implants in vestibule–palatinal and
mesial–distal directions. It is considered that the distance between the implants should be minimum 3 mm in order to prevent potential resorption in bone [29]. Tarnow et al. [29] reported that when the distance between two implants exceeds 3 mm, crestal bone loss is 0.45 mm; when it is less than 3 mm, crestal bone loss is 1.04 mm. It is suggested that a minimum of 1 mm of bone should be left in the vestibule and palatinal surfaces of implants in order to ensure an appropriate exit profile of the implant from the alveolar crest; and for sufficient feeding of the bone. Spray et al. [30] reported that resorptions decreased and even bone apposition was observed when there is 1.8 mm or more of bone in the vestibule surfaces of implants. Bidez and Misch reported that strain in bone was lower in cases with 3 implants when compared to those with 2 implants [31,32]. Considering the above factors and clinical practices, the present study modeled 5 different implant localizations leaving a minimum dis-
442
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 12. Strain distribution and values in the trabecular bone in the Model 2.
tance of 3 mm between the implants and a minimum of 1 mm in vestibule and palatinal regions of implants. There is a large body of research analyzing the amount of stress transferred to dental implants and peri-implant bone by the materials used in implant-supported prostheses [33–37]. Considering previous research and clinical practices, the present study chromium– cobalt alloy (Wiron 99; Bego, Bremen, Germany) [33,38,39] supported feldspastic porcelain (Ceramco II; Dentsply, Burlington, ABD) [33,34,40] were used as implant supported prosthesis. FE studies in oral implantology applied various occlusal forces of 100–2000 N [18,22,27,41–43]. In our study, 100 N mean vertical chewing force was used [42]. There are a limited number of previous studies on loading in different regions of the anterior maxilla. Clelland et al. [22] applied occlusal loads in the direction of the longer axis of implants. Saab et al. [43] applied occlusal loads from the cingulum region, and Hsu et al. [44] applied the occlusal loads at angles of 0°, 30° and 60° to the longer axis of the implant. Forces applied along the longer axis of implants represent the situation when the upper and lower incisors are closed. This type of closing during chewing
is observed in a small proportion of patients with distorted upper– lower jaw relationship. Furthermore, the measured chewing force is lower when the mandible is in eccentric position and the posterior teeth do not contact [43]. The application of vertical loads in buccal and apical direction from the cingulum regions of crowns reflects the location of mandibular incisors closest to the palatinal surfaces of maxillary incisors; in other words, centric occlusion location. Centric occlusion position is the appropriate area in which to make examine the magnitude of vertical loading applied to the anterior maxilla [43]. Therefore, in the present study, vertical chewing forces were applied at a 45° angle from cingulum regions. Force application over dental implants, abutment or crowns affects the reliability of the results. Application of chewing forces to the crowns provides realistic results [44]. For this reason, the present study applied loads to the crowns. It is difficult to make calculations according to permanent deformations in order to analyze an element. Instead, the analysis treats the systems as linear elastic; in other words, it is more convenient to analyze an idealized model of the system. In the present
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
443
Fig. 13. Strain distribution and values in the trabecular bone in the Model 3.
study, the upper jaw was considered as linear elastic for all calculations. Furthermore, the maxilla was considered as homogenous and isotropic, but is not a homogenous structure in practice. Similarly, there are varying opinions on whether the jaws are isotropic or orthotropic [23,45,46]. In a study that modeled the jaw as orthotropic, stress increased by 25%. In this study, osseointegration was assumed to be 100% in order to reduce calculation time. However, osseointegration and stress in peri-implant bones is proportional [47]. Spivey et al. [47] reported that when osseointegration was 83.3% strain measured in peri-implant bone increased by 5% compared to the situation when osseointegration was considered as 100%. One limitation of our study is the fact that Frost’s mechanostat theory was not specifically developed for dental implants. Comprehensive researches are still needed to determine bone strain values
under occlusal loads for peri-implant bone, although Frost himself notified that human load-bearing bones adapt to mechanical changes [7–9]. The most important disadvantage of FEA analyses is that the results are only valid for the selected model. For example, assumptions are made for variable data such as the dimensions of dental implants, cortical bone thickness, trabecular bone density and amount of osseointegration. Therefore, it should be remembered that the results will vary when the input data change. 6. Conclusion This study evaluated biomechanical behavior for implant localizations of fabricating implant retained fixed prostheses, in the anterior maxilla. Our findings reveal that increasing the number of implants
444
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
Fig. 14. Strain distribution and values in the trabecular bone in the Model 4.
Fig. 15. Strain distribution and values in the trabecular bone in the Model 5.
N. Bölükbası, ¸ S. Yeniyol / Medical Engineering and Physics 37 (2015) 431–445
reduced the risk of bone resorption. Different model types should be evaluated and prospective clinical studies with long-term follow-up are required for definitive conclusions. Conflict of interests None declared. Funding None. Ethical approval The use of patient data was carried out according to the policies and procedures of the Istanbul University, Faculty of Dentistry, Department of Oral Implantology. The patient gave written consent to use the CT images for the purposes of this study. References [1] Buser D, Martin W, Belser UC. Optimizing esthetics for implant restorations in the anterio maxilla: anatomic and surgical considerations. Int J Oral Maxillofac Implants 2004;19(suppl):43–61. [2] Belser UC, Buser D, Higginbottom F. Consensus statements and recommended clinical procedures regarding esthetics in implant dentistry. Int J Oral Maxillofac Implants 2004;19(suppl):73–4. [3] Belser UC, Schmid B, Higginbottom F, Buser D. Outcome analysis of implant restorations located in the anterior maxilla: a review of the recent literature. Int J Oral Maxillofac Implants 2004;19(suppl):30–42. [4] Pietrokowski J. The bony residual ridge in man. J Prosthet Dent 1975;34:456–62. [5] Misch CE, Bidez MW. Implant protected occlusion a biomechanical rationale. Compend Cont Educ Dent 1994;15:1330–43. [6] Misch CE. Premaxilla implant considerations: surgery and fixed prosthodontics. In: Misch CE, editor. Contemporary implant dentistry. St Louis: Mosby; 1993. p. 575–97. [7] Stanford CM, Brand RA. Toward an understanding of implant occlusion and strain adaptive bone modelling and remodelling. J Prosthet Dent 1999;81:553–61. [8] Fung YC. Biomechanics: Mechanical properties of living tissues. 3rd ed. New York: Springer- Verlag; 1993. p. 1–58. [9] Frost HM. Perspectives: bone’s mechanical usage windows. Bone Miner 1992;19:257–71. [10] Frost HM. A 2003 update of bone physiolology and Wollf’s law for clinicians. Angle Orthod 2004;74:3–15. [11] Isidor F. Occlusal loading in implant dentistry. In: Lang NP, Karring T, Lindhe J, editors. Proceedings of the third European workshop on periodontology—implant dentistry. Berlin: Quientessence Books; 1999. p. 358–75. [12] Bilgin T. Research concerning about the optimum prefabricated impression trays for total edentulous patients in Turkey [Ph.D. thesis]. Istanbul, Turkey: Istanbul University; 1989. [13] Ash MM. Wheeler’s dental anatomy, physiology and occlusion. Philadelphia, PA: WB Saunders; 1993. p. 128–273. [14] Lekholm U, Zarb GA. Patient selection and preparation. In: Branemark P-I, Zarb GA, Albrektsson T, editors. Tissue integrated prostheses: osseointegration in clinical dentistry. Chicago, IL: Quintessence; 1985. p. 199–209. [15] Sevimay M, Turhan F, Kılıçaslan MA, Eskitasçıoglu G. Three-dimensional finite element analysis of the different bone quality on stress distribution in an implantsupported crown. J Prosthet Dent 2005;93(3):227–34. [16] Himmlova L, Dostolava T, Kacovsky A, Konvickova S. Influence of implant length and diameter on stress distribution: a finite element analysis. J Prosthet Dent 2004;91(1):20–5. [17] I˙ plikçiolu H, Akça K. Comparative evaluation of the effect of diameter, length and number of implants supporting three-unit fixed partial prostheses on stress distribution in the bone. J Dent 2002;30(1):41–6. [18] Çaglar A, Aydın C, Yılmaz C, Korkmaz T. Effects of mesio distal inclination of implants on stress distribution in implant-supported fixed prostheses. Int J Oral Maxillofac Implants 2006;21:36–44. [19] Tepper G, Haas R, Zechner W, Krach W, Watzek G. Three-dimensional finite element analysis of implant stability in the atrophic posterior maxilla: a mathematical study of the sinus floor augmentation. Clin Oral Implants Res 2002;13:657–65.
445
[20] Lin CL, Wang JC, Ramp LC, Liu PR. Biomechanical response of implant systems placed in the maxillary posterior region under various conditions of angulation, bone density, and loading. Int J Oral Maxillofac Implants 2008;23:57–64. [21] Okumura N, Stegaroiu R, Nishiyama H, Kurokawa K, Kitamura E, Hayashi T, et al. Finite element analysis of implant-embedded maxilla model from CT data: comparison with the conventional model. J Prosthodont Res 2011;55:24–31. [22] Clelland NL, Lee KL, Bimbenet OC, Brantley WA. A three-dimensional finite element stress analysis of angled abutments for an implant placed in the anterior maxilla. J Prosthodont 1995;4:95–100. [23] Clelland NL, Lee JK, Bimbenet OC, Gilat A. Use of an axisymmetric finite element method to compare maxillary bone variables for a loaded implant. J Prosthodont 1993;2:183–9. [24] Sano M, Ikebe K, Yang TC, Maeda Y. Biomechanical rationale for six splinted implants in bilateral canine, premolar, and molar regions in an edentulous maxilla. Implant Dent 2012;21:220–4. [25] Lee JS, Lim YJ. Three-dimensional numerical simulation of stress induced by different lengths of osseointegrated implants in the anterior maxilla. Comput Methods Biomech Biomed Eng 2013;16(11):1143–9. [26] Sadrimanesh R, Siadat H, Sadr-Eshkevari P, Monzavi A, Maurer P, Rashad A. Alveolar bone stress around implants with different abutment angulation: an FEanalysis of anterior maxilla. Implant Dent 2012;21:196–201. [27] Sagat G, Yalcin S, Gultekin BA, Mijiritsky E. Influence of arch shape and implant position on stress distribution around implants supporting fixed full-arch prosthesis in edentulous maxilla. Implant Dent 2010;19:498–508. [28] Tian K, Chen J, Han L, Yang J, Huang W, Wu D. Angled abutments result in increased or decreased stress on surrounding bone of single-unit dental implants: a finite element analysis. Med Eng Phys 2012;34:1526–31. [29] Tarnow DP, Cho SC, Wallace SS. The effect of inter-implant distance on the height of inter-implant bone crest. J Periodontol 2000;71:546–9. [30] Spray JR, Black CG, Morris HF, Ochi S. The influence of bone thickness on facial marginal bone response: stage 1 placement through stage 2 uncovering. Ann Periodontol 2000;5:119–28. [31] Bidez MW, Misch CE. The biomechanics of inter-implant spacing; 1990. [32] Bidez MW, Misch CE. Issues in bone mechanics related to oral implants. Implant Dent 1992;1:289–94. [33] Sevimay M, Üsümez A, Eskitasçıoglu G. The influence of various occlusal materials on stresses transferred to implant-supported prostheses and supporting bone: a three-dimensional finite-element study. J Biomed Mater Res Part B Appl Biomater 2005;73:140–7. [34] Cibirka RM, Razzoog ME, Lang BR, Stohler CS. Determining the force absorption quotient for restorative materials used in implant occlusal surfaces. J Prosthet Dent 1992;67:361–4. [35] Eskitascıoglu G, Baran I, Aykac Y, Oztas D. Investigation of the effect of different esthetic materials in implant-crown design. Turkish J Oral Imp 1996;4:13–19. [36] Sertgöz A. Finite element analysis study of the effect of superstructure material on stress distribution in implant supported fixed prosthesis. Int J Prosthodont 1997;10:19–27. [37] Papavasiliou G, Kamposiora P, Bayne SC, Felton DA. Three dimensional finite element analysis of stress distribution around single tooth implants as a function of bony support, prosthesis type and loading during function. J Prosthet Dent 1996;76:633–40. [38] Wi1lliams KR, Watson CJ, Murphy WM, Scott J, Gregory M, Sinobad D. Finite element analysis of fixed prostheses attached to osseointegrated implants. Quintessence Int 1990;21:563–70. [39] O’Brien WJ. Dental materials and their selection. 2nd ed. Chicago: Quintessence; 1997. p. 259–72. [40] Hojjatie B, Anusavice KJ. Three-dimensional finite element analysis of glassceramic dental crowns. J Biomech 1990;23:1157–66. [41] Bozkaya D, Müftü S, Müftü A. Evaluation of load transfer characteristics of five different implants in compact bone at different load levels by finite elements analysis. J Prosthet Dent 2004;9:523–30. [42] Tada S, Stegaroiu R, Kitamura E, Miyakawa O, Kusakari H. Influence of implant design and bone quality on stress/strain distribution in bone around implants: a 3dimensional finite element analysis. Int J Oral Maxillofac Implants 2003;18:357– 68. [43] Saab XE, Griggs JA, Powers JM, Engelmeier RL. Effect of abutment angulation on the strain on the bone around an implant in the anterior maxilla: a finite element study. J Prosthet Dent 2007;97:85–92. [44] Hsu ML, Chen FC, Kao HC, Cheng CK. Influence of off-axis loading of an anterior maxillary implant: a three-dimensional finite element analysis. Int J Oral Maxillofac Implants 2007;22:301–9. [45] Ashman RB, Van Buskirk WC. The elastic properties of a human mandible. Adv Dent Res 1987;1:64–7. [46] Cowin SC. Bone mechanics. Boca Raton, FL: CRC; 1989. p. 27–141. [47] Spivey JD, Kong W, Fotos PG. Stress distribution at the bone-to-implant interface. a 3-D finite element analyses (Abstract 106). J Dent Res 1993;72:117.
