All-ceramic crowns have been introduced in

Transcription

1 Influence of Core Thickness on a Restored Crown of a First Premolar Using Finite Element Analysis Kaarel A. Proos, BEng, PhD a Michael V. Swain, BSc, PhD b Jim Ironside, BDS, MDS, PhD c Grant P. Steven, BSc, PhD d Purpose: This study examined the influence of coping thickness on the maximum stresses that arise in a first premolar all- crown. Materials and Methods: Axisymmetric finite element models with different In-Ceram Alumina coping thicknesses (0.3, 0.6, and 0.9 mm) were examined. Models with and without resin lute were constructed. To all models, an identical axial load of 600 N was applied vertically downward, over an area around the crown s fissure. Results: The resulting peak tensile maximum principal stresses in each part of the crown existed below the fracture strengths of the respective materials making up the crown. This was true for all variations of core thickness, with and without resin lute. The peak tensile stresses in the coping, porcelain, and dentin decreased for an increase in core thickness. This was most evident in the porcelain and coping. Conclusion: The thickness of the core has a significant influence on the resulting stresses in the coping, porcelain, and dentin of this axially loaded crown. Int J Prosthodont 2003;16: All- crowns have been introduced in prosthodontics for better esthetics with acceptable strength. In contrast to metal- restorations, light can be transmitted through the restoration and underlying tooth structure to ensure an improvement in translucency. 1 In the construction of all crown restorations, the relationship between the different layers of material that make up the a Postdoctoral Fellow, School of Aerospace, Mechanical and Mechatronic Engineering and Faculty of Dentistry, University of Sydney, Biomaterials Science Research Unit, National Innovation Centre, Eveleigh, New South Wales, Australia. b Professor, School of Aerospace, Mechanical and Mechatronic Engineering and Faculty of Dentistry, University of Sydney, Biomaterials Science Research Unit, National Innovation Centre, Eveleigh, New South Wales, Australia. c Doctor, Faculty of Dentistry, University of Sydney, Biomaterials Science Research Unit, National Innovation Centre, Eveleigh, New South Wales, Australia. d Professor, School of Engineering, University of Durham, United Kingdom. Reprint requests: Dr Michael V. Swain, School of Aerospace, Mechanical and Mechatronic Engineering and Faculty of Dentistry, University of Sydney, Biomaterials Science Research Unit, Suite G11, National Innovation Centre, Australian Technology Park, Eveleigh, New South Wales 1430, Australia. Fax: mswain@mail.usyd.edu.au restoration specifically the ratio of thickness between porcelain, coping, and dentin is not clearly understood. Such an understanding is necessary for the restoration to be fully functional and reliable. Many factors influence the interplay between the necessary thicknesses of these layers, including fracture strength (critical loads), 2 5 flexural strength, 5 7 stress distribution, 5,8 modulus of rupture, 9 translucency, 1,10 and shade matching. 11,12 It is known that the relatively stiff coping plays a significant role in bearing the load of the restoration. 2,5,13 Different fracture tests have been conducted on multilayer systems to identify this role more clearly. Hertzian contact testing 2 on monolayer, bilayer (simulating monolithic crowns on dentin), and trilayer (simulating veneer/core all- crowns on dentin) structures provides an ideal starting point to understand the fundamental mechanics of failure in layered dental s under loading conditions that represent basic occlusal function. Also, contact fracture testing of trilayer structures (complemented with finite element analysis [FEA]) has found that the fracture resistance of trilayer structures (imitating porcelain veneered on a stiff core supported on a soft substrate) appears to be more dependent on the core thickness than on the veneer thickness. 5 The International Journal of Prosthodontics 474

2 Proos et al Influence of Core Thickness on a Crown by FEA Fig 1 Vertical cross-section and finite element mesh of first premolar with thickness. The models with 0.3- and 0.9- mm core thicknesses had similar cross-sections and an identical mesh. 600 N 0.5-mm-radius line angle Coping 0.6-mm shoulder radius 0.7 mm 0.7 mm 2.5 mm 2.5 mm Pulp cavity 5.0 mm Fixed mm mm 2.0 mm mm Table 1 Properties of Dental Materials Used Young s Poisson s Fracture Material modulus (E, GPa) ratio ( ) strength (MPa) In-Ceram Alumina Adhesive resin* *Bond strength to dentin = 11 MPa. 27 The individuality of crown shapes both externally and internally further complicates both the fracture pattern and the resistance to crack initiation. 5 As yet, analytic closed-form relations for simple planar trilayer and complex trilayer structures (simulating a restored crown) have not been developed. It is necessary to move on from simple planar analyses to more realistic multilayer analyses that incorporate relevant geometric features such as specimen surface curvature. 2 Evaluating the stress distribution in restored crowns for different loading configurations will assist in this. The present study concludes a series of studies by examining how the stresses inside a first premolar all- crown are influenced as a result of varying the thickness of the coping. The hypothesis was that the core thickness would have an influence on the resulting stresses of the axially loaded crown. This article seeks to identify the relationship between core thickness and the maximum stresses generated, using FEA on an axisymmetric model. FEA studies have not yet been comprehensively done to examine the effect of coping thickness on the resulting stresses in restored crowns. This has been done for metal- restorations Materials and Methods For the present study, the effects of three different core thicknesses were examined. Three different axisymmetric finite element models were constructed, each with a different core thickness (0.3, 0.6, and 0.9 mm). All models contained an In-Ceram Alumina core (Vita), veneered with a layer of porcelain that was luted to the underlying dentin with adhesive resin cement (0.1 mm thick). Another three models similar to the first were also constructed, but without any lute. Here, it was assumed that the coping was bonded directly to the dentin. An identical finite element mesh was used for each model (Fig 1) with a different coping thickness. These models were constructed by switching the material properties (Table 1) of particular elements to make up the premolar crown of specified core thickness. Specifically, a Poisson s ratio ( ) of 0.3 was selected from the range to for feldspathic porcelain. Each model was axisymmetric in fashion, where it was assumed that the shape of every radial slice of the tooth was identical. This was done as in the 475 The International Journal of Prosthodontics

3 Influence of Core Thickness on a Crown by FEA Proos et al 11 peak = MPa 11 peak = MPa 11 peak = MPa Plate stress: 11 MPa a b c Fig 2 11 contour plot of crown for different core thicknesses: a = 0.3 mm; b = 0.6 mm; c = 0.9 mm. TT peak = MPa TT peak = 92.2 MPa TT peak = 76.7 MPa Plate stress: TT MPa a b c Fig 3 Hoop stress ( TT ) contour plot of crown for different core thicknesses: a = 0.3 mm; b = 0.6 mm; c = 0.9 mm. preceding series of studies Similar to these, a downward vertical edge stress of 71.3 MPa was applied to the model, imitating a surface pressure on top of the crown. This was the equivalent of applying a high bite force 27 of 600 N over an area of radius 1.64 mm. Also, the models were completely restrained at the base, where any translation or rotation of nodes was not permitted. Each model was constructed with 3,745 two-dimensional quadrilateral and triangular elements. More specific modeling details, such as the anatomic dimensions of the crown and modeling assumptions, are outlined in the preceding studies Results Stress contour plots of the models with three different core thicknesses and with resin lute are given in Figs 2 and 3. Figure 2 displays the maximum principal stresses ( 11 ), and Fig 3 displays the hoop stresses ( TT ), which exist in a circular fashion around the axis of symmetry. Figures 4 to 6 convey the 11 across various sections of the crown for all three models containing cement. A comparison of the peak tensile 11 in the coping, porcelain, dentin, and cement is displayed in Table 2. The International Journal of Prosthodontics 476

4 Proos et al Influence of Core Thickness on a Crown by FEA Maximum principal stress (MPa) A A B Vertical distance from fissure (mm) B Fig 4 11 distribution of each core thickness system versus vertical distance measured directly beneath and parallel to the direction of the applied load of the crown from fissure (A) along the axis of symmetry to a depth of 1.6 mm below fissure (B). Maximum principal stress (MPa) C C D Horizontal distance from axis of symmetry (mm) Fig 5 11 distribution of each core thickness system versus horizontal distance from axis of symmetry (0.1 mm below coping, ie, C) to crown s outer perimeter (D). D Discussion As in the other studies, the peak tensile 11 was located on the inside of the coping, in the occlusal area. The peak tensile 11 in the porcelain existed on the surface of the crown, partially outside the crown s cusp, and the peak tensile 11 in the dentin was located directly beneath the position where the chamfer margin began turning upward. A large proportion of these peak tensile 11 were due to the hoop stresses ( TT ). Comparing the magnitude of the highest 11 with the corresponding fracture strengths of the matching material, it is obvious that for the differing coping thicknesses, the stresses did not exceed the respective strengths. This was true for both models with and without resin lute. It was also found that the shear tensile stress values (radial/axial plane) in the lute were significantly below the bond strength of adhesive resin to dentin by approximately 80%. Note that these evaluations of minimum likelihood of fracture are for this specific axisymmetric model with this specific axial load case. Figures 5 and 6 support the idea 2,5,13 that the coping plays a crucial role in bearing the applied load 477 The International Journal of Prosthodontics

5 Influence of Core Thickness on a Crown by FEA Proos et al Maximum principal stress (MPa) E E F F Distance from axis of symmetry (mm) Fig 6 11 distribution from a radius of 1.5 mm from the axis of symmetry (E) to the tooth perimeter near the margin (F) for the different core thickness systems. Height: 0.5 mm above cervical edge. Table 2 Comparison of Peak Tensile 11 (MPa) for Each Core Thickness with and without ation* Core Peak 11 in coping Peak 11 in porcelain Peak 11 in dentin Peak 11 thickness (mm) ation No cementation ation No cementation ation No cementation in cement *Peak tensile 11 in the lute are not valid for the crowns not modeling the resin lute. on the crown. This is obvious from the high 11 seen in the core- compared to the 11 of the other materials. The significance of the stiff coping is also obvious from the fact that the 11 values in the coping are greatly sensitive to a coping thickness change. It is thus necessary to determine what influence the increase in coping thickness will have on the maximum stresses. By increasing the coping thickness by 100% (from 0.3 to 0.6 mm), the highest peak tensile stresses in the coping, porcelain, and dentin decreased by 23%, 42%, and 3%, respectively. By increasing the coping thickness another 50% (from 0.6 to 0.9 mm), the stresses decreased by another 15%, 49%, and 0%, respectively. The 11 values in the coping were significantly sensitive to a coping thickness change. What was more evident is the sensitivity of the 11 in the porcelain to this same thickness change. These stresses are much more responsive than those found in another FEA study 18 ; the tensile stress in porcelain (at the porcelain margin) increased by 14% for a decrease in Ni-Cr core thickness from 0.3 to 0.1 mm and increased by 4% for a decrease in the same thickness of an Au-Pd core. In that study, a force of 200 N was distributed near the metal-porcelain junction. In another study of metal- systems, 19 a reduction in the coping thickness had a negligible effect on the stress levels in porcelain. These studies, however, have a different coping type (ie, metal coping), models of anatomy, dimensions, and load cases. Also, the percentage change in coping thickness studied was different. Concerning the fracture resistance of all- crowns with 0.5- and 0.7-mm-thick alumina cores, it was found that no statistical difference existed between them. 4 In contrast, investigations 5,6 into the role of core and porcelain thickness found that the mean flexural strength of disks tested biaxially is significantly greater for the core material compared to a bilayer disk containing core and porcelain of the same thickness. It was recommended that lamination be avoided in areas where maximum strength is required in In-Ceram all- crowns and fixed partial dentures. In an examination 28 of the strength of all- crowns fabricated by the Procera system (Nobel Biocare), it was recommended that the thickness of the coping material be maximized and the thickness of the veneering porcelain be minimized. The International Journal of Prosthodontics 478

6 Proos et al Influence of Core Thickness on a Crown by FEA Obviously, there are many different results concerning the influence of coping thickness. Because these fracture tests are not purely numeric (as with FEA, which assumes homogenous, isotropic material with no voids present), other factors, such as material strength, 2 7 toughness, 29 and Weibull modulus, affect the influence of core thickness. Direct comparisons between empiric and numeric results are not possible because of the different materials involved and their associated behavior. Nevertheless, it can still be surmised from this and the above studies that the coping thickness significantly affects the peak tensile 11, particularly in the porcelain and coping. Specifically, a thicker core produces lower peak tensile 11 in the coping, porcelain, and dentin. Such a reason to increase the core thickness, however, is limited because a critical requirement for the clinical use of restored crowns is that they be esthetically the same as natural teeth. A drawback with the use of glass-infiltrated alumina is the reflective interface at the junction between the veneer porcelain and the alumina core. If the core is too close to the surface, the crown no longer looks natural. 5 There are concerns of the porcelain/core thickness associated with influences on translucency and shade matching. 11,12 Also, with regard to the dentin, it is common knowledge that tooth removal needs to be as conservative as possible (ie, atraumatic preparation). Guidelines for scientific tooth preparation have been outlined in a recent article. 33 Conclusion From the specific axisymmetric model with the axial load applied over the fissure of the all- crown, it may be concluded that increasing the thickness of the relatively stiff coping will decrease the peak tensile 11 in the coping, porcelain veneer, and dentin. These stress reductions are quite significant in the coping and even more so in the porcelain. The implication of such results is that from a numeric analysis viewpoint, it is desirable to increase the thickness of the coping. However, the increase in this coping thickness is limited so that the esthetics of the crown are not compromised and tooth removal of the underlying dentin remains as conservative as possible. Acknowledgments The authors gratefully acknowledge Vita Zahnfabrik, Bad Säckingen, Germany, and the Australian Dental Research Foundation for supporting the project. They also thank Dr Bruce Pardey, Dr Linny Angker, Dr Koji Yamada, Dr Massimiliano Guazzato, and Mr Mohammad Albakry for their technical advice. References 1. Antonson SA, Anusavice KJ. Contrast ratio of veneering and core s as a function of thickness. Int J Prosthodont 2001;14: Lawn BR, Deng Y, Thompson VP. Use of contact testing in the characterization and design of all- crownlike layer structures: A review. J Prosthet Dent 2001;86: Hopkins K. An investigation into the role of porcelain thickness in determining the load-carrying capacity of porcelain laminates. Br Dent J 1989;167: Abed HM, Razzoog ME, Lang BR, Yaman P. The effect of alumina core thickness on the fracture resistance of all- crowns [abstract 394]. J Dent Res 1997;76: Ironside J. An In Vitro Analysis of the Behaviour of an Alumina Based Dental All-Ceramic Restorative System Subjected to Occlusal Loads [thesis]. Sydney: The University of Sydney, AlShehri S. An investigation into the role of core porcelain thickness and lamination in determining the flexural strength of InCeram system [abstract 18]. J Dent Res 1999;78: Wagner WC, Chu TM. Apparent flexural strength of porcelain veneered all- core material [abstract 2129]. J Dent Res 1996; 75: Proos KA, Steven G, Swain MV, Ironside J. Preliminary studies on the optimum shape of dental bridges. Comp Meth Biomech Biomed Eng 2000;4: White SN, Caputo AA, Li ZC, Zhao XY. Modulus of rupture of the Procera all- system. J Esthet Dent 1996;8: Odén A, Razzoog ME. Masking ability of Procera AllCeram copings of various thickness [abstract 2376]. J Dent Res 1997;76: Douglas RD, Przybylska M. Predicting porcelain thickness required for dental shade matches. J Prosthet Dent 1999;82: Przybylska M, Douglas D. Predicting porcelain thickness for dental shade matches [abstract 336]. J Dent Res 1997;76: Proos KA, Swain MV, Ironside J, Steven G. Finite element analysis studies of an all- crown on a first premolar. Int J Prosthodont 2002;15: Proos KA, Swain MV, Ironside J, Steven G. Finite element analysis studies of a metal- crown on a first premolar. Int J Prosthodont 2002;15: Proos KA, Swain MV, Ironside J, Steven G. Influence of cement on a restored crown of a first premolar using finite element analysis. Int J Prosthodont 2003;16: Proos KA, Swain MV, Ironside J, Steven G. Influence of margin design and taper abutment angle on a restored crown of a first premolar using finite element analysis. Int J Prosthodont 2003;16: Magne P, Versluis A, Douglas WH. Rationalization of incisor shape: Experimental-numerical analysis. J Prosthet Dent 1999;81: Anusavice KJ, Hojjatie B. Stress distribution in metal- crowns with a facial porcelain margin. J Dent Res 1987;66: Anusavice KJ, Hojjatie B, DeHoff PH. Influence of metal thickness on stress distribution in metal- crowns. J Dent Res 1986; 65: Kamposiora P, Papavasiliou G, Bayne S, Felton D. Stress concentration in all- posterior fixed partial dentures. Quintessence Int 1996;27: Papvasiliou G, Tripodakis A, Kamposiora P, Strub J, Bayne S. Finite element analysis of abutment-restoration combinations for osseointegrated implants. Int J Prosthodont 1996;9: Huysmans M, Van der Varst P. Finite element analysis of quasistatic and fatigue failure of posts and cores. J Dent 1993;21: The International Journal of Prosthodontics

7 Influence of Core Thickness on a Crown by FEA Proos et al 23. Magne P, Douglas W. Optimization of resilience and stress distribution in porcelain veneers for the treatment of crown-fractured incisors. Int J Periodontics Restorative Dent 1999;19: Combe E. Notes on Dental Materials. London: Churchill Livingstone, 1986: Chu G. Glass-s as a new dental porcelain Its properties and potential applications. In: Yamada H (ed). Dental : The State of the Art Los Angeles: University of Southern California, 1977: Guazzato M, Albakry M, Swain MV, Ironside JG. Mechanical properties of InCeram Alumina and InCeram Zirconia [abstract 0911]. J Dent Res 2001;80: Craig RG, Powers JM. Restorative Dental Materials, ed 11. St Louis: Mosby, Andersson M, Razzoog ME, Odén A, Hegenbarth EA, Lang BR. Procera: A new way to achieve an all- crown. Quintessence Int 1998;29: Fischer H, Marx R. Fracture toughness of dental s: Comparison of bending and indentation method. Dent Mater 2002;18: Kelly JR. Perspectives on strength. Dent Mater 1995;11: Tesk JA, Anusavice KJ. Summary of Conference on Design of Dental Prostheses. Dent Mater 1988;4: Hornberger H. Strength Microstructure Relationships in a Dental Alumina Glass Composite [thesis]. Birmingham, UK: The University of Birmingham, 1995: Goodacre CJ, Campagni WV, Aquilino SA. Tooth preparations for complete crowns: An art form based on scientific principles. J Prosthet Dent 2001;85: Literature Abstract The retention of complete crowns prepared with three different tapers and luted with four different cements. The authors studied the retention of crowns with three levels of taper luted with two conventional and two adhesive resin luting agents. There were 120 sound human molars randomly assigned to three groups (n = 40). Each group received complete-crown preparations with tapers of 6, 12, and 24 degrees. Custom-made high noble alloy crowns were fabricated for each preparation. The crowns were cemented with one of the following luting agents: zinc phosphate, conventional glass-ionomer, and two resin cements (C&B Metabond and Panavia). The cemented crowns were separated from the prepared teeth under tension to measure retention. With each cement group, the 6-degree taper was used as a control. The effect of tapers and cement type on the retention of the crowns was tested with ANOVA. Failure modes were also recorded. The results showed: (1) there was a significant difference in the cement and taper; (2) the retention of crowns prepared with 6-degree taper was not significantly different from the 12-degree taper; (3) there was a significant difference in retention between taper at 6 and 24 degrees and between 12 and 24 degrees; (4) the mean retention values of the two adhesive resin cements were significantly higher than those of the zinc phosphate and glass-ionomer cements; and (5) 65% of the failures were adhesive in the luting agent, 31% were cohesive in the tooth, and 4% were assembly failure (fracture of embedding resin). The mode of failure was correlated to taper and cement type. Retention of the adhesive resin cements at 24 degrees of taper was 20% higher than the conventional cements at 6 degrees of taper. Conventional cements (zinc phosphate or glassionomer) yielded retention that was only 50% of the values of the resin cements tested. The result of this study may indicate the routine usage of adhesive resin cements as luting agents for crowns instead of conventional cements in any degree of taper in tooth preparation. Zidan O, Ferguson GC. J Prosthet Dent 2003;89: References: 29. Reprints: Dr Omar Zidan, School of Dentistry, University of Minnesota, Moos Tower, 515 Delaware Street SE, Minneapolis, Minnesota Fax: + (612) zidano@umn.edu Ansgar C. Cheng, Toronto The International Journal of Prosthodontics 480