- Email: [email protected]
Design of a custom angled abutment for dental implants using computer-aided design and nonlinear finite element analysis
ARTICLE IN PRESS Journal of Biomechanics 43 (2010) 1941–1946
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Design of a custom angled abutment for dental implants using computer-aided design and nonlinear finite element analysis Ting Wu a,n, Wenhe Liao a, Ning Dai a, Chunbo Tang b a b
College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao St., Nanjing 210016, PR China Department of Dental Implantology, Jiangsu Provincial Stomatological Hospital, Nanjing 210029, PR China
a r t i c l e in fo
abstract
Article history: Accepted 11 March 2010
Computer-aided design/computer-aided manufacturing (CAD/CAM) custom abutments have been attracting more and more attention due to their advantages of accuracy fit and esthetic emergence profile. However, the CAD key technology for custom abutments has been seldom studied as well as their biomechanical behavior. This paper explored a novel method to design a CAD/CAM custom angled abutment, evaluated the biomechanical performance of the whole system and compared the difference between the custom and the conventional abutment through 3D nonlinear finite element analysis (FEA). Firstly, the digital data of the dental casts at the healing abutment level was acquired by optical scanner. Thus the position of the healing abutment and the implant can be determined by CAD technology. The custom angled abutment was then designed according to the need of restoration and esthetics with CAD software. The described system can eliminate wax and cast, create an esthetic anatomical emergence profile and provide a satisfactory angle correction. Simulation results indicate that there was no distinct difference in the stress distribution and magnitude of implant-bone interface and screw using the custom or the conventional angled abutment. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Dental implant Abutment Custom Computer–aided design (CAD) Finite element analysis (FEA)
1. Introduction The achievement of an esthetic implant-supported restoration is a constant challenge to restorative dentists. Because of the circular shape of the implant and its smaller diameter, when compared with the root of a natural tooth, a dilemma inevitably occurs of how to construct an artificial crown and abutment system that imitates the natural tooth form (Bichacho, 1998). In the past decades, many abutment designs, such as the UCLA abutment (Lewis et al., 1992) and the DIA anatomic abutment (Daftary, 1995), have been introduced in attempts to resolve the stated complications. However, problems of miscasting, inaccuracy fit and unfavorable antirotational capability cannot be avoided when using these abutments. With the rapid development of computer technology, computeraided design/computer-aided manufacturing (CAD/CAM) custom abutments have been gradually used in some dental implant systems. In comparison to cast abutments, there is no need for complicated and time-consuming titanium casting and postcasting manipulation that may produce inaccuracies; hence, the CAD/CAM abutment fit may be more precise. Moreover, the CAD/ CAM custom abutments have better physical properties since the material is processed from a homogenous mass under more
n
Corresponding author. Tel.: + 86 25 84895881; fax: + 86 25 84891701. E-mail address: [email protected] (T. Wu).
0021-9290/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2010.03.017
controlled conditions, with no need for abutment inventory (Grossmann et al., 2006). The Procera custom abutments (Nobel Biocare, Goteborg, Sweden) produced from Procera copy milling system were found to have many advantages such as ideal emergence profile and machined fit to the implant (Marchack, 1996). But a complicated waxing procedure must be completed to build the abutment to full anatomic contour. Another example is the Encodes Restorative System (3i Implant Innovations, Palm Beach Gardens, Florida) which can eliminate wax and cast. The key to this system is the codes that are embedded into the occlusal surfaces of the Encode healing abutments. Laser optical scanner interprets these codes and the custom abutment is then designed with special CAD software (Drago, 2006; Grossmann et al., 2006). As there are notches on the healing abutment, this Encode system is only limited to 3i implant system. Moreover, the neck of the custom abutment, based on the emergence profile of the Encode healing abutment, is circular shape in cross-section; therefore this abutment is not a real esthetic abutment. To date, the CAD key technology for custom abutments has been seldom studied as well as their biomechanical behaviors. The purpose of this paper is to design a custom angled abutment with esthetic emergence profile by using CAD technology, to evaluate the biomechanical performance of the whole system and to compare the difference between the custom abutment and the conventional abutment through 3D nonlinear finite element analysis (FEA).
ARTICLE IN PRESS 1942
T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
2. Materials and methods 2.1. CAD for the custom angled abutment 2.1.1. Healing abutment-level casts The first step to design a custom abutment is to get the implant position. To avoid implant-level impression, the position of the healing abutment was utilized in this study. The collar height of the selected healing abutment must be at least 1 mm higher than the measured tissue height. After the tissue healing completely, definitive impressions of the healing abutment and the opposing arch were taken and poured to obtain the healing abutment-level dental casts (Fig. 1). 2.1.2. Numerical models An optical scanner was used to scan the dental casts to acquire the numerical models. The final 3D STL data of the dental casts would provide good information for the next design. In order to generate the precise 3D solid models of the implant (F4.1 10 mm, ITI Straumann AG, Waldenburg, Switzerland) and the healing abutment (M2.0 4.5 mm, ITI Straumann AG, Waldenburg, Switzerland) selected in this case, the OGP video measuring system (Optical Gaging Products, Inc., USA) was used to get the detailed shape parameters. Thus all the components can be modeled accurately by the CAD software Pro/Engineer Wildfire 3.0. 2.1.3. Positioning the implant By means of a surface fitting tool, the triangular facets in the healing abutment region from the STL models of the dental cast were fitted with parametric surfaces (Fig. 2a): A: lateral region - cylinder surface B: top region - spherical surface
Fig. 1. Healing abutment-level casts: (a)Healing abutment in place and (b)prepared casts to scan.
Then the axis of the cylinder surface and the center of the sphere surface were extracted to be the datum axis and point for the accurate positioning of the healing abutment. Thus the 3D solid model of the healing abutment can be easily positioned on the cast models according to the datum axis and point. Finally the implant model was seated on the healing abutment model on the basis of their connection surfaces.
2.1.4. Designing the custom abutment To get a functional and esthetic result, a virtual prosthesis was selected from our database of artificial teeth. Considering the centric balanced occlusion, the prosthesis was preliminarily positioned and probed for interference using the ‘‘Collision Detection’’ function based on the ‘‘Least Square method’’. After appropriate distorting, moving and rotating, the final prosthesis fixed in the required position is shown in Fig. 2b. The implant in this case showed a buccal inclination obviously (Fig. 2c), so the abutment angle must be determined to correct the direction of the implant. In this study, the long axis of the virtual prosthesis was served as the reference direction. According to the anatomically morphological characteristics, the middle sagittal plane and coronal plane for the virtual prosthesis were created. Then the intersecting line of these two planes was extracted and regarded as the abutment long axis (Fig. 2d). The angle measured between the abutment long axis and the implant long axis was about 22.51. In order for the abutment to have an anatomical emergence profile, the intersecting line of the prosthesis surface and gingival surface was extracted and regarded as the gingival margin. The ‘‘B-Splines’’ function (Interactively creates a B-spline interpolating curve from the specified points and with the designated constraints.) was used to create the cervical line of the custom abutment to correlate with the gingival margin that rises from the buccal or lingual to the proximal. The cervical line has eight knots that can be adjusted in occlusal plane direction and sagittal plane direction. Two knots determine the mesio-distal width, two knots determine the lingu-buccal width, and the other four knots control the contour shape (Fig. 2e). Considering the shrinkage of the soft tissue, the cervical line of the abutment was set 0.5–2 mm under the individual gingival margin so as to satisfy the basic requests for the esthetics and cleaning. The collar shape of the abutment was controlled by four profiles (Fig. 2f). Two profiles control the mesio-distal shape, and two additional profiles control the bucco-lingual shape. This collar design can allow a maximum tissue volume and stability of the gingiva. The design of the abutment also follows the natural shape of a prepared tooth. And a 1 mm wide rounded shoulder was used for the marginal design. The final 3D solid model of the custom abutment is shown in Fig. 2g. Compared with the conventional angled abutment (Fig. 2h), this custom abutment had a satisfactory anatomical emergence profile. And on the integrated system, the custom abutment was in good fitness and provided a favorable angular correction (Fig. 2i).
Fig. 2. CAD for the custom angled abutment: (a)surface fitting; (b)virtual prosthesis; (c)inclination of the implant; (d)middle sagittal and coronal plane of the virtual prosthesis; (e)design cervical line; (f)design collar shape; (g)custom angled abutment; (h)conventional angled abutment and (i)the integrated system with the final custom angled abutment.
ARTICLE IN PRESS T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
1943
2.2. Finite element analysis
Two different FE models were used to compare the difference (Fig. 3):
2.2.1. Finite element modeling In order to validate the structure of this custom abutment and to study whether it would be beneficial to the stability of bone and implant, finite element method was used to evaluate the biomechanical performance of the whole implant system in the surrounding bone. The geometry of the bone took the shape of the maxillary first premolar region of this patient CT images through image segmentation and spline reconstruction. The cortical bone was modeled with a layer thickness of about 1.5 mm around the implant.
Model 1: Custom angled abutment, implant and abutment screw in the bone Model 2: Conventional angled abutment, implant and abutment screw in the bone
Model 1
2.2.2. Materials properties In this study the mechanical properties of implant, abutment and screw were all treated to be isotropic, homogeneous and linear elastic. And titanium alloy (Ti– 6Al–4V) with Young’s modulus of 110 GPa and Poisson’s ratio of 0.33 was used for the simulation of these components (Geramy and Morgano, 2004). Bone is an anisotropic material, which means it has different physical properties when measured in different directions. Recent studies have concluded that anisotropy has significant effects on peri-implant stress and strain and careful consideration should be given to its use in biomechanical finite elements studies (Liao et al., 2008). For this reason, the anisotropy of cortical and cancellous bone was used in this study. Their elastic material constants are presented in Table 1.
Model 2
Buccal
All the solid models were transferred to ANSYS Workbench 11.0 (Swanson Analysis Inc., Huston, PA, USA) to generate the FE models by using 10-node tetrahedral elements. And 0.5 mm of element size was utilized as the meshing requirement for all the FE models.
Lingual
Cortical bone Cancellous bone Implant Screw Abutment Fig. 3. Finite element models of Model 1(custom angled abutment design) and Model 2(conventional angled abutment design).
2.2.3. Boundary and loading conditions The mesial and distal surfaces of the bone segment were constrained in x, y and z directions (displacements ¼ 0) as the boundary condition. Static load of 150 N was applied on the top surface of the abutment along the abutment long axis and along the implant long axis to simulate two different loading conditions, respectively. Moreover, nonlinear contact zones were defined at implant-bone, implantabutment, implant-screw and abutment-screw interfaces. Contact analysis assured the transfer of the load and deformation between the different components. The coefficient of friction was set 0.3 between all the titanium– titanium interfaces (Alkan et al., 2004). The 0.65 friction coefficient was selected
Table 1 Elastic coefficients for cortical and cancellous bone used in the study.
Corticala Cancellousb
E1(GPa)
E2(GPa)
E3(GPa)
G12(GPa)
G23(GPa)
G13(GPa)
v12
v23
v13
12.7 0.21
17.9 1.148
22.8 1.148
5.0 0.068
7.4 0.434
5.5 0.068
0.18 0.055
0.28 0.322
0.31 0.055
a Elastic coefficients for cortical bone are based on Schwartz-Dabney and Dechow (2003). The one-direction is radial, the two-direction is tangential (circumferential), and the three-direction is axial (longitudinal). b Elastic coefficients for cancellous bone are based on O’Mahony et al. (2000). The one-direction is infero-superior (the axis of transverse isotropy symmetry with the smallest of Young’s modulus value), the two-direction is medial–lateral, and the three-direction is anterior–posterior.
Fig. 4. Von-Mises stress distribution of implant, custom abutment, conventional abutment and screw under loading along the abutment long axis (row 1) and along the implant long axis (row 2).
ARTICLE IN PRESS 1944
T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
for the cortical bone-implant interface (Yu et al., 2005), and the value of 0.77(Grant et al., 2007) for the cancellous bone-implant interface.
3. Results For the abutment, implant and screw, von-Mises stress (EQV) may be important to interpret the stress occurred, so their vonMises stress distributions are presented in Fig. 4. While the bone can be classified as brittle and anisotropic material, therefore principle stress is appropriate to analyze such a work (Liao et al., 2008). In detail, if it is assumed as a physiological limit state that overload occurs when ultimate bone strength is reached, it follows that maximum principal compressive and tensile stress moduli on cortical bone should be less than 170–190 MPa and 100–130 MPa, respectively, whereas normal stress modulus on cancellous bone (both in compression and tension) should be less than about 5 MPa (Baggi et al., 2008). For the illustrated contour patterns in Figs. 5 and 6, the tensile and compression values are in positive and negative signs respectively to have a conventional engineering representation. 3.1. Loading along the abutment long axis The simulated results (Table 2) demonstrate that when loading along the abutment long axis, the maximum von-Mises stress value of the custom abutment (86.9 MPa) was about 30% lower than that of the conventional abutment (122.6 MPa). The former’s maximum stresses were located at the bottom that contacted with abutment screw, while the latter’s maximum stresses were found at the access hole for straining screw. There was no obvious difference between Model 1 and Model 2 in the stress distribution and magnitude of implant, abutment screw and bone. High stresses of the implant were mainly concentrated at the shoulder and
Fig. 6. Distribution of maximum principal stress and minimum principal stress in cancellous bone under loading along the abutment long axis (row 1) and along the implant long axis (row 2).
Fig. 5. Distribution of maximum principal stress and minimum principal stress in cortical bone under loading along the abutment long axis (row 1) and along the implant long axis (row 2).
Table 2 Maximum Von-Mises stress (EQV) in abutment, implant and screw, and maximum principal tensile stress (S1) and minimum principal compressive stress (S3) in bone of the two FE models under the two loading conditions Component
Abutment (EQV) Implant (EQV) Screw (EQV) Cortical bone (S1) Cortical bone (S3) Cancellous bone (S1) Cancellous bone (S3)
Loading along the abutment long axis
Loading along the implant long axis
Model 1 (custom)
Model 2 (conventional)
Model 1 (custom)
Model 2 (conventional)
86.9 113.5 41.8 20.4 48.1 3.0 3.0
122.6 111.7 40.8 20.4 48.1 3.0 3.0
323.9 360.9 91.6 19.5 62.8 5.2 4.4
308.5 396.1 93.7 19.5 62.8 5.2 4.4
ARTICLE IN PRESS T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
implant-abutment joint on the buccal side. The maximum stresses of the screw were observed at the first thread on the buccal side. For the cortical bone, peak compression (48.1 MPa) occurred at the crest around implant neck on the buccal side, and peak tension (20.4 MPa) at the junction between cortical and cancellous bone on the buccal side. For the cancellous bone, peak tension and compressive stresses were both found near the first thread groove. 3.2. Loading along the implant long axis Loading along the implant long axis remarkably raised stresses of almost all the components due to the bending stress contributed by the transverse force component. Although the custom abutment possessed a 5% higher maximum stress than the conventional one, the implant stress decreased by 9% in custom abutment case. The maximum stresses of the custom abutment and the conventional one were both concentrated on the buccal side of the bottom. High stresses of the implant were mainly localized at the shoulder and implant-abutment joint on the lingual side. There was no significant difference between Model 1 and Model 2 in the stress distribution and magnitude of screw and bone. The maximum stresses of the screw were observed at the first thread on the lingual side. For the cortical bone, peak compression (62.8 MPa) occurred at the crest on the lingual side, and peak tension (19.5 MPa) at the junction between cortical and cancellous bone on the lingual side. Loading along the implant long axis increased the peak compressive stress of cortical bone by 30% as compared with loading along the abutment axis. For the cancellous bone, peak tension and compressive stresses were both found near the first thread groove on the lingual side. Moreover, the peak tension stress value (5.2 MPa) exceeded the physiological limit (5 MPa), but most stress was still within the physiologic range.
4. Discussion Until the present time, few studies on CAD/CAM custom abutments have been done in the literature especially on the design process and biomechanical mechanism of this treatment modality. The present study explored a novel method to design a custom angled abutment, avoiding wax and cast, offering an esthetic anatomical emergence profile and providing a satisfactory angle correction. Through 3D nonlinear FEA, the biomechanical performance of this abutment was evaluated and compared with that of the conventional abutment. It might provide the clues to design a custom angled abutment and interpret the biomechanical effect of its clinical use. For CAD of a custom abutment, it is important to record the position of the implant in 3 dimensions. Procera CAD/CAM technique uses a screw with a graduated pin for determining height of the abutment placed in the replica of the implant on the master cast to allow the technician to visually align the computer image with the master cast. So it is impossible to relay exactly the implant position from the master cast to the computer (Kucey and Fraser, 2000). In this study, we easily got the implant position through fitting the triangular mesh of the healing abutment to parametric surface. The method for correcting the implant angle is directly correlated to the restorative accuracy. However, we cannot find definite descriptions in the studies before. So the long axis of the virtual prosthesis was served as the reference direction in our study. Based on classic teeth morphology theory, we defined two reference planes including middle sagittal plane and coronal plane for virtual prosthesis according to the anatomical and morphological characteristics. And the results also show that this method corrected the implant angle effectively.
1945
The esthetic appearance is determined to a considerable extend by the gingival contour. The method in this study can conveniently create a satisfactory esthetic emergence profile controlled by a 3D B-Splines curve following the nature shape of the virtual prosthesis. But after adequate maturation of the soft tissue, the gingival margin should be evaluated and the abutment may be adjusted as necessary. Because the emergence profile is created by the abutment and not the restoration, the laboratory procedure can be simplified. The FEA results demonstrate that using the custom abutment could not significantly influence the stress distribution and magnitude of implant-bone interface and screw in both loading conditions. Perhaps the reason is that loading is resisted mainly by the Morse-tapered connection, where a high normal pressure is maintained in the contact area. The tapered interference fit relies on the large contact pressure and resulting frictional resistance to provide a stable connection. Form lock and friction are the basic determinants that affected the magnitude and distribution of stresses. And this mating, referred to as positive or geometric locking, is assumed to be responsible for protecting the abutment screws from excessive load (Akc-a et al., 2003). The two FE Models have the same angulation, the same implant-abutment connection, and the same implant-bone interface; therefore the implant-bone interface of the two Models experienced the same stress under the same external load. The results also indicate that high stresses of the cortical bone, mainly in compression, were located at the crestal region around the implant, which correspond with the clinical finding of maxillary crestal bone loss. Loading along the implant long axis caused higher stresses of almost all the components compared with loading along the abutment axis. And small areas of tension stress higher than 5 MPa were found within the cancellous bone near the first thread groove. But most stress was still within the physiologic range. So it has always been recommended to direct occlusal loads as close to the long axis of the abutments/restorations as possible. However, the real loading condition is mostly at a certain angle towards the long axis of the abutments. The clinical performances of angled abutments have mostly been satisfactory. Further studies are needed to clarify the discrepancies between the favorable clinical success rates and the unfavorable in biomechanical research data (Hsu et al., 2005). In addition, the whole system is still in experimental set-up phase at present, so the evaluation of its reliability and accuracy was preliminary. In the future, laboratory quantitative tests and clinic experiments will be carried out in order to improve and practice the system. Conflict of interest statement The authors declare no competing financial interests.
Acknowledgement This study was supported by the Doctoral fund of Ministry of Education Program of China (Grant No. 20070287055). References Akc- a, K., C - ehreli, M.C., Iplikc- io˘glu, H., 2003. Evaluation of the mechanical characteristics of the implant-abutment complex of a reduced-diameter morse-taper implant. A nonlinear finite element stress analysis. Clinical Oral Implants Research 14, 444–454. Alkan, I., Sertg¨oz, A., Ekici, B., 2004. Influence of occlusal forces on stress distribution in preloaded dental implant screws. The Journal of Prosthetic Dentistry 91, 319–325. Baggi, L., Cappelloni, I., Maceri, F., Vairo, G., 2008. Stress-based performance evaluation of osseointegrated dental implants by finite-element simulation. Simulation Modelling Practice and Theory 16, 971–987.
ARTICLE IN PRESS 1946
T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
Bichacho, N., 1998. Achieving optimal gingival esthetics around restored natural teeth and implants. Rationale, concepts, and techniques. Dental Clinics of North America 42, 763–780. Daftary, F., 1995. The bio-esthetic abutment system: an evolution in implant prosthetics. International Journal of Dental Symposia 3, 10–15. Drago, C.J., 2006. Two new clinical/laboratory protocols for CAD/CAM implant restoration. The Journal of the American Dental Association 137, 794–800. Geramy, A., Morgano, S.M., 2004. Finite element analysis of three designs of an implant-supported molar crown. The Journal of Prosthetic Dentistry 92, 434–440. ¨ Grant, J.A., Bishop, N.E., Gotzen, N., Sprecher, C., Honl, M., Morlock, M.M., 2007. Artificial composite bone as a model of human trabecular bone: the implantbone interface. Journal of Biomechanics 40, 1158–1164. Grossmann, Y., Pasciuta, M., Finger, I.M., 2006. A novel technique using a coded healing abutment for the fabrication of a CAD/CAM titanium abutment for an implant-supported restoration. The Journal of Prosthetic Dentistry 95, 258–261. Hsu, M.L., Chung, T.F., Kao, H.C., 2005. Clinical applications of angled abutments—a literature review. Chinese Dental Journal 24, 15–20.
Kucey, B.K.S., Fraser, D.C., 2000. The Procera abutment—the fifth generation abutment for dental implants. Journal of the Canadian Dental Association 66, 445–449. Lewis, S.G., Llamas, D., Avera, S., 1992. The UCLA abutment: a four-year review. The Journal of Prosthetic Dentistry 67, 509–515. Liao, S.H., Tong, R.F., Dong, J.X., 2008. Influence of anisotropy on peri-implant stress and strain in complete mandible model from CT. Computerized Medical Imaging and Graphics 32, 53–60. Marchack, C.B., 1996. A custom titanium abutment for the anterior single-tooth implant. The Journal of Prosthetic Dentistry 76, 288–291. O’Mahony, A.M., Williams, J.L., Katz, J.O., Spencer, P., 2000. Anisotropic elastic properties of cancellous bone from a human edentulous mandible. Clinical Oral Implants Research 11, 415–421. Schwartz-Dabney, C.L., Dechow, P.C., 2003. Variations in cortical material properties throughout the human dentate mandible. American Journal of Physical Anthropology 120, 252–277. Yu, H.Y., Cai, Z.B., Zhou, Z.R., Zhu, M.H., 2005. Fretting behavior of cortical bone against titanium and its alloy. Wear 259, 910–918.
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Design of a custom angled abutment for dental implants using computer-aided design and nonlinear finite element analysis Ting Wu a,n, Wenhe Liao a, Ning Dai a, Chunbo Tang b a b
College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao St., Nanjing 210016, PR China Department of Dental Implantology, Jiangsu Provincial Stomatological Hospital, Nanjing 210029, PR China
a r t i c l e in fo
abstract
Article history: Accepted 11 March 2010
Computer-aided design/computer-aided manufacturing (CAD/CAM) custom abutments have been attracting more and more attention due to their advantages of accuracy fit and esthetic emergence profile. However, the CAD key technology for custom abutments has been seldom studied as well as their biomechanical behavior. This paper explored a novel method to design a CAD/CAM custom angled abutment, evaluated the biomechanical performance of the whole system and compared the difference between the custom and the conventional abutment through 3D nonlinear finite element analysis (FEA). Firstly, the digital data of the dental casts at the healing abutment level was acquired by optical scanner. Thus the position of the healing abutment and the implant can be determined by CAD technology. The custom angled abutment was then designed according to the need of restoration and esthetics with CAD software. The described system can eliminate wax and cast, create an esthetic anatomical emergence profile and provide a satisfactory angle correction. Simulation results indicate that there was no distinct difference in the stress distribution and magnitude of implant-bone interface and screw using the custom or the conventional angled abutment. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Dental implant Abutment Custom Computer–aided design (CAD) Finite element analysis (FEA)
1. Introduction The achievement of an esthetic implant-supported restoration is a constant challenge to restorative dentists. Because of the circular shape of the implant and its smaller diameter, when compared with the root of a natural tooth, a dilemma inevitably occurs of how to construct an artificial crown and abutment system that imitates the natural tooth form (Bichacho, 1998). In the past decades, many abutment designs, such as the UCLA abutment (Lewis et al., 1992) and the DIA anatomic abutment (Daftary, 1995), have been introduced in attempts to resolve the stated complications. However, problems of miscasting, inaccuracy fit and unfavorable antirotational capability cannot be avoided when using these abutments. With the rapid development of computer technology, computeraided design/computer-aided manufacturing (CAD/CAM) custom abutments have been gradually used in some dental implant systems. In comparison to cast abutments, there is no need for complicated and time-consuming titanium casting and postcasting manipulation that may produce inaccuracies; hence, the CAD/CAM abutment fit may be more precise. Moreover, the CAD/ CAM custom abutments have better physical properties since the material is processed from a homogenous mass under more
n
Corresponding author. Tel.: + 86 25 84895881; fax: + 86 25 84891701. E-mail address: [email protected] (T. Wu).
0021-9290/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2010.03.017
controlled conditions, with no need for abutment inventory (Grossmann et al., 2006). The Procera custom abutments (Nobel Biocare, Goteborg, Sweden) produced from Procera copy milling system were found to have many advantages such as ideal emergence profile and machined fit to the implant (Marchack, 1996). But a complicated waxing procedure must be completed to build the abutment to full anatomic contour. Another example is the Encodes Restorative System (3i Implant Innovations, Palm Beach Gardens, Florida) which can eliminate wax and cast. The key to this system is the codes that are embedded into the occlusal surfaces of the Encode healing abutments. Laser optical scanner interprets these codes and the custom abutment is then designed with special CAD software (Drago, 2006; Grossmann et al., 2006). As there are notches on the healing abutment, this Encode system is only limited to 3i implant system. Moreover, the neck of the custom abutment, based on the emergence profile of the Encode healing abutment, is circular shape in cross-section; therefore this abutment is not a real esthetic abutment. To date, the CAD key technology for custom abutments has been seldom studied as well as their biomechanical behaviors. The purpose of this paper is to design a custom angled abutment with esthetic emergence profile by using CAD technology, to evaluate the biomechanical performance of the whole system and to compare the difference between the custom abutment and the conventional abutment through 3D nonlinear finite element analysis (FEA).
ARTICLE IN PRESS 1942
T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
2. Materials and methods 2.1. CAD for the custom angled abutment 2.1.1. Healing abutment-level casts The first step to design a custom abutment is to get the implant position. To avoid implant-level impression, the position of the healing abutment was utilized in this study. The collar height of the selected healing abutment must be at least 1 mm higher than the measured tissue height. After the tissue healing completely, definitive impressions of the healing abutment and the opposing arch were taken and poured to obtain the healing abutment-level dental casts (Fig. 1). 2.1.2. Numerical models An optical scanner was used to scan the dental casts to acquire the numerical models. The final 3D STL data of the dental casts would provide good information for the next design. In order to generate the precise 3D solid models of the implant (F4.1 10 mm, ITI Straumann AG, Waldenburg, Switzerland) and the healing abutment (M2.0 4.5 mm, ITI Straumann AG, Waldenburg, Switzerland) selected in this case, the OGP video measuring system (Optical Gaging Products, Inc., USA) was used to get the detailed shape parameters. Thus all the components can be modeled accurately by the CAD software Pro/Engineer Wildfire 3.0. 2.1.3. Positioning the implant By means of a surface fitting tool, the triangular facets in the healing abutment region from the STL models of the dental cast were fitted with parametric surfaces (Fig. 2a): A: lateral region - cylinder surface B: top region - spherical surface
Fig. 1. Healing abutment-level casts: (a)Healing abutment in place and (b)prepared casts to scan.
Then the axis of the cylinder surface and the center of the sphere surface were extracted to be the datum axis and point for the accurate positioning of the healing abutment. Thus the 3D solid model of the healing abutment can be easily positioned on the cast models according to the datum axis and point. Finally the implant model was seated on the healing abutment model on the basis of their connection surfaces.
2.1.4. Designing the custom abutment To get a functional and esthetic result, a virtual prosthesis was selected from our database of artificial teeth. Considering the centric balanced occlusion, the prosthesis was preliminarily positioned and probed for interference using the ‘‘Collision Detection’’ function based on the ‘‘Least Square method’’. After appropriate distorting, moving and rotating, the final prosthesis fixed in the required position is shown in Fig. 2b. The implant in this case showed a buccal inclination obviously (Fig. 2c), so the abutment angle must be determined to correct the direction of the implant. In this study, the long axis of the virtual prosthesis was served as the reference direction. According to the anatomically morphological characteristics, the middle sagittal plane and coronal plane for the virtual prosthesis were created. Then the intersecting line of these two planes was extracted and regarded as the abutment long axis (Fig. 2d). The angle measured between the abutment long axis and the implant long axis was about 22.51. In order for the abutment to have an anatomical emergence profile, the intersecting line of the prosthesis surface and gingival surface was extracted and regarded as the gingival margin. The ‘‘B-Splines’’ function (Interactively creates a B-spline interpolating curve from the specified points and with the designated constraints.) was used to create the cervical line of the custom abutment to correlate with the gingival margin that rises from the buccal or lingual to the proximal. The cervical line has eight knots that can be adjusted in occlusal plane direction and sagittal plane direction. Two knots determine the mesio-distal width, two knots determine the lingu-buccal width, and the other four knots control the contour shape (Fig. 2e). Considering the shrinkage of the soft tissue, the cervical line of the abutment was set 0.5–2 mm under the individual gingival margin so as to satisfy the basic requests for the esthetics and cleaning. The collar shape of the abutment was controlled by four profiles (Fig. 2f). Two profiles control the mesio-distal shape, and two additional profiles control the bucco-lingual shape. This collar design can allow a maximum tissue volume and stability of the gingiva. The design of the abutment also follows the natural shape of a prepared tooth. And a 1 mm wide rounded shoulder was used for the marginal design. The final 3D solid model of the custom abutment is shown in Fig. 2g. Compared with the conventional angled abutment (Fig. 2h), this custom abutment had a satisfactory anatomical emergence profile. And on the integrated system, the custom abutment was in good fitness and provided a favorable angular correction (Fig. 2i).
Fig. 2. CAD for the custom angled abutment: (a)surface fitting; (b)virtual prosthesis; (c)inclination of the implant; (d)middle sagittal and coronal plane of the virtual prosthesis; (e)design cervical line; (f)design collar shape; (g)custom angled abutment; (h)conventional angled abutment and (i)the integrated system with the final custom angled abutment.
ARTICLE IN PRESS T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
1943
2.2. Finite element analysis
Two different FE models were used to compare the difference (Fig. 3):
2.2.1. Finite element modeling In order to validate the structure of this custom abutment and to study whether it would be beneficial to the stability of bone and implant, finite element method was used to evaluate the biomechanical performance of the whole implant system in the surrounding bone. The geometry of the bone took the shape of the maxillary first premolar region of this patient CT images through image segmentation and spline reconstruction. The cortical bone was modeled with a layer thickness of about 1.5 mm around the implant.
Model 1: Custom angled abutment, implant and abutment screw in the bone Model 2: Conventional angled abutment, implant and abutment screw in the bone
Model 1
2.2.2. Materials properties In this study the mechanical properties of implant, abutment and screw were all treated to be isotropic, homogeneous and linear elastic. And titanium alloy (Ti– 6Al–4V) with Young’s modulus of 110 GPa and Poisson’s ratio of 0.33 was used for the simulation of these components (Geramy and Morgano, 2004). Bone is an anisotropic material, which means it has different physical properties when measured in different directions. Recent studies have concluded that anisotropy has significant effects on peri-implant stress and strain and careful consideration should be given to its use in biomechanical finite elements studies (Liao et al., 2008). For this reason, the anisotropy of cortical and cancellous bone was used in this study. Their elastic material constants are presented in Table 1.
Model 2
Buccal
All the solid models were transferred to ANSYS Workbench 11.0 (Swanson Analysis Inc., Huston, PA, USA) to generate the FE models by using 10-node tetrahedral elements. And 0.5 mm of element size was utilized as the meshing requirement for all the FE models.
Lingual
Cortical bone Cancellous bone Implant Screw Abutment Fig. 3. Finite element models of Model 1(custom angled abutment design) and Model 2(conventional angled abutment design).
2.2.3. Boundary and loading conditions The mesial and distal surfaces of the bone segment were constrained in x, y and z directions (displacements ¼ 0) as the boundary condition. Static load of 150 N was applied on the top surface of the abutment along the abutment long axis and along the implant long axis to simulate two different loading conditions, respectively. Moreover, nonlinear contact zones were defined at implant-bone, implantabutment, implant-screw and abutment-screw interfaces. Contact analysis assured the transfer of the load and deformation between the different components. The coefficient of friction was set 0.3 between all the titanium– titanium interfaces (Alkan et al., 2004). The 0.65 friction coefficient was selected
Table 1 Elastic coefficients for cortical and cancellous bone used in the study.
Corticala Cancellousb
E1(GPa)
E2(GPa)
E3(GPa)
G12(GPa)
G23(GPa)
G13(GPa)
v12
v23
v13
12.7 0.21
17.9 1.148
22.8 1.148
5.0 0.068
7.4 0.434
5.5 0.068
0.18 0.055
0.28 0.322
0.31 0.055
a Elastic coefficients for cortical bone are based on Schwartz-Dabney and Dechow (2003). The one-direction is radial, the two-direction is tangential (circumferential), and the three-direction is axial (longitudinal). b Elastic coefficients for cancellous bone are based on O’Mahony et al. (2000). The one-direction is infero-superior (the axis of transverse isotropy symmetry with the smallest of Young’s modulus value), the two-direction is medial–lateral, and the three-direction is anterior–posterior.
Fig. 4. Von-Mises stress distribution of implant, custom abutment, conventional abutment and screw under loading along the abutment long axis (row 1) and along the implant long axis (row 2).
ARTICLE IN PRESS 1944
T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
for the cortical bone-implant interface (Yu et al., 2005), and the value of 0.77(Grant et al., 2007) for the cancellous bone-implant interface.
3. Results For the abutment, implant and screw, von-Mises stress (EQV) may be important to interpret the stress occurred, so their vonMises stress distributions are presented in Fig. 4. While the bone can be classified as brittle and anisotropic material, therefore principle stress is appropriate to analyze such a work (Liao et al., 2008). In detail, if it is assumed as a physiological limit state that overload occurs when ultimate bone strength is reached, it follows that maximum principal compressive and tensile stress moduli on cortical bone should be less than 170–190 MPa and 100–130 MPa, respectively, whereas normal stress modulus on cancellous bone (both in compression and tension) should be less than about 5 MPa (Baggi et al., 2008). For the illustrated contour patterns in Figs. 5 and 6, the tensile and compression values are in positive and negative signs respectively to have a conventional engineering representation. 3.1. Loading along the abutment long axis The simulated results (Table 2) demonstrate that when loading along the abutment long axis, the maximum von-Mises stress value of the custom abutment (86.9 MPa) was about 30% lower than that of the conventional abutment (122.6 MPa). The former’s maximum stresses were located at the bottom that contacted with abutment screw, while the latter’s maximum stresses were found at the access hole for straining screw. There was no obvious difference between Model 1 and Model 2 in the stress distribution and magnitude of implant, abutment screw and bone. High stresses of the implant were mainly concentrated at the shoulder and
Fig. 6. Distribution of maximum principal stress and minimum principal stress in cancellous bone under loading along the abutment long axis (row 1) and along the implant long axis (row 2).
Fig. 5. Distribution of maximum principal stress and minimum principal stress in cortical bone under loading along the abutment long axis (row 1) and along the implant long axis (row 2).
Table 2 Maximum Von-Mises stress (EQV) in abutment, implant and screw, and maximum principal tensile stress (S1) and minimum principal compressive stress (S3) in bone of the two FE models under the two loading conditions Component
Abutment (EQV) Implant (EQV) Screw (EQV) Cortical bone (S1) Cortical bone (S3) Cancellous bone (S1) Cancellous bone (S3)
Loading along the abutment long axis
Loading along the implant long axis
Model 1 (custom)
Model 2 (conventional)
Model 1 (custom)
Model 2 (conventional)
86.9 113.5 41.8 20.4 48.1 3.0 3.0
122.6 111.7 40.8 20.4 48.1 3.0 3.0
323.9 360.9 91.6 19.5 62.8 5.2 4.4
308.5 396.1 93.7 19.5 62.8 5.2 4.4
ARTICLE IN PRESS T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
implant-abutment joint on the buccal side. The maximum stresses of the screw were observed at the first thread on the buccal side. For the cortical bone, peak compression (48.1 MPa) occurred at the crest around implant neck on the buccal side, and peak tension (20.4 MPa) at the junction between cortical and cancellous bone on the buccal side. For the cancellous bone, peak tension and compressive stresses were both found near the first thread groove. 3.2. Loading along the implant long axis Loading along the implant long axis remarkably raised stresses of almost all the components due to the bending stress contributed by the transverse force component. Although the custom abutment possessed a 5% higher maximum stress than the conventional one, the implant stress decreased by 9% in custom abutment case. The maximum stresses of the custom abutment and the conventional one were both concentrated on the buccal side of the bottom. High stresses of the implant were mainly localized at the shoulder and implant-abutment joint on the lingual side. There was no significant difference between Model 1 and Model 2 in the stress distribution and magnitude of screw and bone. The maximum stresses of the screw were observed at the first thread on the lingual side. For the cortical bone, peak compression (62.8 MPa) occurred at the crest on the lingual side, and peak tension (19.5 MPa) at the junction between cortical and cancellous bone on the lingual side. Loading along the implant long axis increased the peak compressive stress of cortical bone by 30% as compared with loading along the abutment axis. For the cancellous bone, peak tension and compressive stresses were both found near the first thread groove on the lingual side. Moreover, the peak tension stress value (5.2 MPa) exceeded the physiological limit (5 MPa), but most stress was still within the physiologic range.
4. Discussion Until the present time, few studies on CAD/CAM custom abutments have been done in the literature especially on the design process and biomechanical mechanism of this treatment modality. The present study explored a novel method to design a custom angled abutment, avoiding wax and cast, offering an esthetic anatomical emergence profile and providing a satisfactory angle correction. Through 3D nonlinear FEA, the biomechanical performance of this abutment was evaluated and compared with that of the conventional abutment. It might provide the clues to design a custom angled abutment and interpret the biomechanical effect of its clinical use. For CAD of a custom abutment, it is important to record the position of the implant in 3 dimensions. Procera CAD/CAM technique uses a screw with a graduated pin for determining height of the abutment placed in the replica of the implant on the master cast to allow the technician to visually align the computer image with the master cast. So it is impossible to relay exactly the implant position from the master cast to the computer (Kucey and Fraser, 2000). In this study, we easily got the implant position through fitting the triangular mesh of the healing abutment to parametric surface. The method for correcting the implant angle is directly correlated to the restorative accuracy. However, we cannot find definite descriptions in the studies before. So the long axis of the virtual prosthesis was served as the reference direction in our study. Based on classic teeth morphology theory, we defined two reference planes including middle sagittal plane and coronal plane for virtual prosthesis according to the anatomical and morphological characteristics. And the results also show that this method corrected the implant angle effectively.
1945
The esthetic appearance is determined to a considerable extend by the gingival contour. The method in this study can conveniently create a satisfactory esthetic emergence profile controlled by a 3D B-Splines curve following the nature shape of the virtual prosthesis. But after adequate maturation of the soft tissue, the gingival margin should be evaluated and the abutment may be adjusted as necessary. Because the emergence profile is created by the abutment and not the restoration, the laboratory procedure can be simplified. The FEA results demonstrate that using the custom abutment could not significantly influence the stress distribution and magnitude of implant-bone interface and screw in both loading conditions. Perhaps the reason is that loading is resisted mainly by the Morse-tapered connection, where a high normal pressure is maintained in the contact area. The tapered interference fit relies on the large contact pressure and resulting frictional resistance to provide a stable connection. Form lock and friction are the basic determinants that affected the magnitude and distribution of stresses. And this mating, referred to as positive or geometric locking, is assumed to be responsible for protecting the abutment screws from excessive load (Akc-a et al., 2003). The two FE Models have the same angulation, the same implant-abutment connection, and the same implant-bone interface; therefore the implant-bone interface of the two Models experienced the same stress under the same external load. The results also indicate that high stresses of the cortical bone, mainly in compression, were located at the crestal region around the implant, which correspond with the clinical finding of maxillary crestal bone loss. Loading along the implant long axis caused higher stresses of almost all the components compared with loading along the abutment axis. And small areas of tension stress higher than 5 MPa were found within the cancellous bone near the first thread groove. But most stress was still within the physiologic range. So it has always been recommended to direct occlusal loads as close to the long axis of the abutments/restorations as possible. However, the real loading condition is mostly at a certain angle towards the long axis of the abutments. The clinical performances of angled abutments have mostly been satisfactory. Further studies are needed to clarify the discrepancies between the favorable clinical success rates and the unfavorable in biomechanical research data (Hsu et al., 2005). In addition, the whole system is still in experimental set-up phase at present, so the evaluation of its reliability and accuracy was preliminary. In the future, laboratory quantitative tests and clinic experiments will be carried out in order to improve and practice the system. Conflict of interest statement The authors declare no competing financial interests.
Acknowledgement This study was supported by the Doctoral fund of Ministry of Education Program of China (Grant No. 20070287055). References Akc- a, K., C - ehreli, M.C., Iplikc- io˘glu, H., 2003. Evaluation of the mechanical characteristics of the implant-abutment complex of a reduced-diameter morse-taper implant. A nonlinear finite element stress analysis. Clinical Oral Implants Research 14, 444–454. Alkan, I., Sertg¨oz, A., Ekici, B., 2004. Influence of occlusal forces on stress distribution in preloaded dental implant screws. The Journal of Prosthetic Dentistry 91, 319–325. Baggi, L., Cappelloni, I., Maceri, F., Vairo, G., 2008. Stress-based performance evaluation of osseointegrated dental implants by finite-element simulation. Simulation Modelling Practice and Theory 16, 971–987.
ARTICLE IN PRESS 1946
T. Wu et al. / Journal of Biomechanics 43 (2010) 1941–1946
Bichacho, N., 1998. Achieving optimal gingival esthetics around restored natural teeth and implants. Rationale, concepts, and techniques. Dental Clinics of North America 42, 763–780. Daftary, F., 1995. The bio-esthetic abutment system: an evolution in implant prosthetics. International Journal of Dental Symposia 3, 10–15. Drago, C.J., 2006. Two new clinical/laboratory protocols for CAD/CAM implant restoration. The Journal of the American Dental Association 137, 794–800. Geramy, A., Morgano, S.M., 2004. Finite element analysis of three designs of an implant-supported molar crown. The Journal of Prosthetic Dentistry 92, 434–440. ¨ Grant, J.A., Bishop, N.E., Gotzen, N., Sprecher, C., Honl, M., Morlock, M.M., 2007. Artificial composite bone as a model of human trabecular bone: the implantbone interface. Journal of Biomechanics 40, 1158–1164. Grossmann, Y., Pasciuta, M., Finger, I.M., 2006. A novel technique using a coded healing abutment for the fabrication of a CAD/CAM titanium abutment for an implant-supported restoration. The Journal of Prosthetic Dentistry 95, 258–261. Hsu, M.L., Chung, T.F., Kao, H.C., 2005. Clinical applications of angled abutments—a literature review. Chinese Dental Journal 24, 15–20.
Kucey, B.K.S., Fraser, D.C., 2000. The Procera abutment—the fifth generation abutment for dental implants. Journal of the Canadian Dental Association 66, 445–449. Lewis, S.G., Llamas, D., Avera, S., 1992. The UCLA abutment: a four-year review. The Journal of Prosthetic Dentistry 67, 509–515. Liao, S.H., Tong, R.F., Dong, J.X., 2008. Influence of anisotropy on peri-implant stress and strain in complete mandible model from CT. Computerized Medical Imaging and Graphics 32, 53–60. Marchack, C.B., 1996. A custom titanium abutment for the anterior single-tooth implant. The Journal of Prosthetic Dentistry 76, 288–291. O’Mahony, A.M., Williams, J.L., Katz, J.O., Spencer, P., 2000. Anisotropic elastic properties of cancellous bone from a human edentulous mandible. Clinical Oral Implants Research 11, 415–421. Schwartz-Dabney, C.L., Dechow, P.C., 2003. Variations in cortical material properties throughout the human dentate mandible. American Journal of Physical Anthropology 120, 252–277. Yu, H.Y., Cai, Z.B., Zhou, Z.R., Zhu, M.H., 2005. Fretting behavior of cortical bone against titanium and its alloy. Wear 259, 910–918.