This article was downloaded by: [UGR-BTCA Gral Universitaria] On: 21 March 2012, At: 06:43 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Computer Methods in Biomechanics and Biomedical Engineering Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/gcmb20
Stress transfer properties of different commercial dental implants: a finite element study a
b
a
M. A. Pérez , J. C. Prados-Frutos , J. A. Bea & M. Doblaré a
c d
Aragón Institute of Engineering Research (I3A), University of Zaragoza, Zaragoza, Spain
b
Departamento de Estomatología y Anatomía Humana, Facultad de Ciencias de la Salud, Universidad Rey Juan Carlos, Alcorcón, Madrid, Spain c
Group of Structural Mechanics and Material, Modelling, Aragón Institute of Engineering Research (13A), University of Zaragoza, Zaragoza, Spain d
CIBER-BBN Centro de Investigación en Red en Bioingeniería, Biomateriales y Nanomedicina, Zaragoza, Spain Available online: 24 Jun 2011
To cite this article: M. A. Pérez, J. C. Prados-Frutos, J. A. Bea & M. Doblaré (2012): Stress transfer properties of different commercial dental implants: a finite element study, Computer Methods in Biomechanics and Biomedical Engineering, 15:3, 263-273 To link to this article: http://dx.doi.org/10.1080/10255842.2010.527834
PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.
Computer Methods in Biomechanics and Biomedical Engineering Vol. 15, No. 3, March 2012, 263–273
Stress transfer properties of different commercial dental implants: a finite element study M.A. Pe´reza*, J.C. Prados-Frutosb, J.A. Beaa and M. Doblare´c,d a Arago´n Institute of Engineering Research (I3A), University of Zaragoza, Zaragoza, Spain; bDepartamento de Estomatologı´a y Anatomı´a Humana, Facultad de Ciencias de la Salud, Universidad Rey Juan Carlos, Alcorco´n, Madrid, Spain; cGroup of Structural Mechanics and Material Modelling, Arago´n Institute of Engineering Research (13A), University of Zaragoza, Zaragoza, Spain; dCIBER-BBN Centro de Investigacio´n en Red en Bioingenierı´a, Biomateriales y Nanomedicina, Zaragoza, Spain
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
(Received 15 December 2009; final version received 25 September 2010) Dental implantology has high success rates, and a suitable estimation of how stresses are transferred to the surrounding bone sheds insight into the correct design of implant features. In this study, we estimate stress transfer properties of four commercial implants (GMI, Lifecore, Intri and Avinent) that differ significantly in macroscopic geometry. Detailed three-dimensional finite element models were adopted to analyse the behaviour of the bone-implant system depending on the geometry of the implant (two different diameters) and the bone–implant interface condition. Occlusal static forces were applied and their effects on the bone, implant and bone–implant interface were evaluated. Large diameters avoided overload-induced bone resorption. Higher stresses were obtained with a debonded bone–implant interface. Relative micromotions at the bone– implant interface were within the limits required to achieve a good osseointegration. We anticipate that the methodology proposed may be a useful tool for a quantitative and qualitative comparison between different commercial dental implants. Keywords: finite element analysis; dental implant; bone – implant interface; geometry; stress transfer
1.
Introduction
Since the late 1960s, dental implants have been extensively used for the rehabilitation of completely and partially edentulous patients (Branemark et al. 1969; Adell et al. 1990; Albrektsson and Wennerberg 2005). Longterm success rates as high as 95% for mandibular and maxillary implants have been reported (Jaffin et al. 2004; Ostman et al. 2008; Pikner 2008). Despite these high success rates, early or late implant failure is still reported (Ostman et al. 2008). A key factor for the success or failure of a dental implant is to ensure that the implant can support biting forces and is safely transferred to the interfacial tissues in the long term (Brunski 1992). Overloading of an implant may result in marginal bone resorption of surrounding bone (Quirynen et al. 1992; Carter et al. 1996). Load transfer from implants to surrounding bone depends on the type of loading, the material properties of the implant and prosthesis, the nature of the bone – implant interface, the implant geometry, length, diameter, shape and characteristics of its surface, and the quantity and quality of the surrounding bone (Holmes and Loftus 1997; Geng et al. 2001; Sahin et al. 2002; Bozkaya et al. 2004; Eskitascioglu et al. 2004). Some of these biomechanical issues may be evaluated by means of numerical methods. In recent years, threedimensional (3D) finite element analysis (FEA) has been widely used in applied dentistry for analysing different
*Corresponding author. Email:
[email protected] ISSN 1025-5842 print/ISSN 1476-8259 online q 2012 Taylor & Francis http://dx.doi.org/10.1080/10255842.2010.527834 http://www.tandfonline.com
restorative techniques (Asmussen et al. 2005; Maceri et al. 2007) and implant applications (Geng et al. 2001; van Staden et al. 2006), investigating the influence of implant and prosthesis design (Holmgren et al. 1998; Chun et al. 2002; Himmlova´ et al. 2004; Petrie and Williams 2005; Baggi et al. 2008), magnitude and direction of loads (Holmgren et al. 1998; Alkan et al. 2004; Bozkaya et al. 2004; Chun et al. 2006), bone mechanical properties (van Oosterwyck et al. 1998; Kitagawa et al. 2005; Sevimay et al. 2005) and different bone –implant interface conditions (van Oosterwyck et al. 1998). Holmgren et al. (1998) applied a 2D FE method to determine the effect of implant diameter and different implant design concepts. They concluded that the largest implant diameter will cause minimal trabecular bone stress and that stepped cylindrical implants better distributed the stress than straight cylindrical implants. Recently, Himmlova´ et al. (2004) and Ding et al. (2009) performed a similar study analysing the implant length and diameter. Similar conclusions were obtained, although the implant shape was simplified to a plain cylinder and the bone shape to a prism (Himmlova´ et al. 2004). This fact greatly limits their results. van Oosterwyck et al. (1998) analysed numerically the effect of two extreme bone – implant interface conditions (fixed bond vs. frictionlessfree contact) by means of a 2D FE model. For both interface conditions, great differences in stress values were noticed. In a more recent study, Chun et al. (2005) performed a 3D FEA of the stress distribution in maxillary bone with
M.A. Pe´rez et al.
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
264
different boundary conditions at the interface. They concluded that special care must be taken to assign boundary conditions at the interface for the analysis. van Oosterwyck et al. (1998) and Chun et al. (2005) did not consider the micromovements at the bone – implant interface when debonded. These relative displacements may condition the osseointegration process; in fact, some authors have shown that 50 mm is the upper limit of the relative displacement under which osseointegration takes place at the bone – implant interface (Kienapfel et al. 1999), although some others raise this limit to 200 mm (Miura et al. 1998; Pe´rez del Palomar et al. 2005). Nowadays, advanced modelling of bone – implant interface is being developed assuming its evolutive behaviour and considering other biological events (Bu¨cher et al. 2003; Moreo et al. 2007; Pe´rez and Prendergast 2007; Perez et al. 2008; Moreo et al. 2009). In this paper, it is hypothesised that four commercial implants differ significantly in their mechanical behaviour (stress and relative micromotions) at the bone–implant interface. To test this hypothesis, four implants GMI (Ilerimplant, Lleida, Spain), Lifecore (Lifecore Biomedical, Chasca, MN, USA), Intri (Intri, Zaragoza, Spain) and Avinent (Avinent Implant System, Barcelona, Spain) were analysed with two different diameters (3.75 and 5 mm) and were placed into the first premolar region of the mandible. Their stress transfer properties and relative movement at the bone–implant interface were evaluated, analysing, mainly their diameter and the bone–implant interface condition. They were modelled using a 3D FE technique. For the analysis, two extreme bone–implant interface conditions (bonded and debonded without friction) were studied. 2.
Materials and methods
The first step of the FEA was to represent the geometry of interest, in this case, the implants and the surrounding bone. The four commercial implants were scanned in a Micro-computerised tomography (CT) (GE Locus, GE Healthcare, London, Ontario, Canada) and their geometry was produced as a stereolithography file (Figure 1). Two different diameters of each implant were analysed (3.75 and 5.0 mm). Their length was 10 mm, except for the Intri implants the length of which was 9 mm. Additional details about their geometry are summarised in Table 1. Then, starting from CT data, the first premolar region of a human mandible was modelled using Mimics (Materialise, Leuven, Belgium). Semi-automatic segmentation was used for the region of the mandible to distinguish between cortical and trabecular bone. From the segmentation, a bone block, 26 mm high and 18 mm width, represented the section of the premolar region (Figure 2(a)). The dimensions of the anatomical site portion assumed in the model were set in order to ensure enough distance between the implants and the ends of the model, thus avoiding any
undesired boundary effects. Then, each implant was positioned in that region of the human mandible using Mimics. To allow the load application at the same height with respect to the bone (Bozkaya et al. 2004), a distance of 9.5 mm above the bone was considered as the point for load application (Figures 1 and 2(a)). Abutment and implant were treated as one component. The FE meshes were generated for the different models by an automatic mesher (Harpoon Sharc Ltd, Manchester, UK). A fine FE mesh was used to represent the model of the implants and the bone. In general, increasingly fine mesh size ensures convergence of an FE solution (Cook et al. 2001). Use of a large number of elements was especially important in this problem, where stress singularities were expected at the sharp corners (Figure 1) of the solution domain. Mean value of the mesh size was set equal to approximately 0.2 mm. The number of elements and nodes used in this study for all the models was approximately 680,000 and 450,000, respectively. Table 2 summarises the number of elements and nodes for the convergent discrete models. Implants and bone were modelled with linear, elastic, isotropic and homogeneous properties. Implants were made from titanium alloy (Ti6Al4V). The Young modulus and Poisson ratio of the titanium alloy was 100 GPa and 0.32, respectively (Bidez and Misch 1992). The cortical bone was modelled with a Young modulus of 13.7 GPa, whereas the trabecular bone had a Young modulus of 1.37 GPa (Bozkaya et al. 2004). The Poisson ratio was 0.3 for cortical and trabecular bone (Menicucci et al. 2002; Figure 2(b)). To analyse the stress transfer across the bone, it is important to know its limits of behaviour. The ultimate stress of cortical bone has been reported to be higher in compression (170 MPa) than in tension (100 MPa) (Martin et al. 1998). The strength of the trabecular bone has been reported to be the same in tension and compression and is approximately 2 –5 MPa (Martin et al. 1998). Loading of the implants, in 3D, with forces of 17.1, 114.6 and 23.4 N in a lingual, an axial and a mesiodistal direction, respectively, simulated average masticatory force in a natural, oblique direction (Figure 2(a)). These components represented a masticatory force of 118.2 N in the angle of approximately 758 to the occlusal plane (Mericske-Stern and Zarb 1996). This 3D loading should act on the lingual inclination of the buccal cusp of the crown. Oblique occlusal forces represent a more realistic situation of the dental implant environment (Holmgren et al. 1998). The FE models had the displacements fixed to one row of nodes at the surface of the mandibular region (rotations allowed), and the mesial and distal borders of the end of the mandible section were constrained so that the displacement of nodes in the direction perpendicular to the surface was equal to zero (Figure 2(a)). Two extreme situations were modelled for the bone – implant interface (Figure 2(b)): bonded and debonded without friction
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
Computer Methods in Biomechanics and Biomedical Engineering
265
Figure 1. Geometry of the dental implants analysed in this study (L, implant total length; l, bone – implant interface length; d, implant diameter; p, average thread pitch and t, average thread depth). Load application was at a distance of 9.5 mm above the bone.
diameters and the bone – implant interface condition. The relative motion between bone and implant was also evaluated when the bone –implant interface was debonded. Micromotion may be related to tissue disruption and nonstability of dental implants. The relative differences (in percentage) among designs were also computed taking the design with the lowest stress/micromotion as the reference and then normalise all other values to this (Rel. Dif (%)).
(van Oosterwyck et al. 1998). Full bone contact was assumed between bone and implant surface. To define the contact between bone and implant (debonded without friction), non-penetration constraints were automatically imposed by Abaqus. Implant and bone are both defined as (deformable) bodies. All boundary nodes from one body (e.g. the implant) will be prevented from crossing the boundary surface of another body (e.g. the bone). All the analyses were performed with the commercial FE software package Abaqus v.6.5. (Simulia, Suresnes, France). The stress field on the implant (von Mises stress) and surrounding bone (maximum principal stress) was evaluated for the case of previous static loading distinguishing between the different implant designs and Table 1.
Results
The von Mises stress distributions predicted for the commercial dental implants (bonded bone – implant interface) are represented in Figure 3. The highest
Geometric properties of the commercial implants analysed in this study.
d (mm) GMI GMI Lifecore Lifecore Intri Intri Avinent Avinent
3.
3.75 5 3.75 5 3.75 5 3.75 5
L (mm)
l (mm)
p (mm)
t (mm)
Angle of convergence (8)
10 10 10 10 9 9 10 10
8.5 8.4 8.5 8.5 7.6 7.8 8.4 8.8
0.80 0.85 0.62 0.80 0.77 0.80 0.61 0.81
0.54 0.65 0.26 0.51 0.36 0.39 0.33 0.44
5 5 5 5 5 5 5 5
Bone– implant interface surface (mm2) 111.8 143.1 116.1 155.3 89.9 103.1 123.1 152.1
Notation refers to Figure 1: L, implant total length; l, bone – implant interface length; d, implant diameter; p, average thread pitch and t, average thread depth.
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
266
M.A. Pe´rez et al.
Figure 2. (a) Applied loads and boundary conditions in FE model (B, buccal; D, distal; L, lingual and M, mesial). The height of load application is also indicated. (b) Regions modelled in the FE model of dental implants and mandibular bone.
von Mises stress values were localised around the neck of the implants varying from about 22.38 MPa (Imp Avinent – 5 mm) to 83.30 MPa (Imp GMI – 3.75 mm). These values were much lower than the yield strength of the titanium alloy (650 MPa). Furthermore, stresses in the dental implants were dissipated from the upper screw threads to the implant apex by ensuring that the extreme apical zone barely received stress (Figure 3). When analysing stress levels in the simulated implants, their diameter is a decisive factor in the reduction in masticatory stress. When the diameter was increased from 3.75 to 5 mm, more homogeneous stress distribution and reduced stress concentration were obtained (Figure 3). The GMI implants showed that the maximum von Mises stresses were reduced from 83.30 to 43.45 MPa when the diameter was increased, which supposed a stress reduction
of 48%. Similar results could be observed for other dental implant families (Lifecore, Intri, and Avinent). von Mises stress values doubled (in percentage) when the implant diameter changed from 5 to 3.75 mm (see relative Table 2. Number of nodes and elements used in finite element models of the commercial implants. Diameter (mm) GMI GMI Lifecore Lifecore Intri Intri Avinent Avinent
3.75 5 3.75 5 3.75 5 3.75 5
Nodes
Elements
441.341 443.647 469.143 483.882 417.356 462.227 468.387 482.236
652.381 716.676 690.847 709.404 648.798 660.726 698.231 707.501
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
Computer Methods in Biomechanics and Biomedical Engineering
267
Figure 3. von Mises stress (MPa) in dental implants for bonded bone – implant interface (D, distal and M, mesial) (Rel. Dif. (%) refers to relative differences (in percentage) among designs, it was computed taking the design with the lowest stresses as the reference and then normalised all other values to this). The black straight line indicates the marginal bone level.
differences among designs in Figure 3). However, the influence of the implant length on stress distributions is less significant. The Intri and Lifecore implants (5 mm), with lengths of 9 and 10 mm (respectively), had maximum von Mises stresses of 32.63 and 27.77 MPa, respectively. A slight reduction was observed when the length was increased. Lifecore (5 mm) and Avinent (5 mm) implants predicted the lowest relative differences; this fact could be related to their similar geometrical characteristics (Table 1). The maximum principal stresses in the mandibular bone for the bonded bone – implant interface are represented in Figure 4. Maximum principal stress levels predicted in the cortical bone, ranging from 8 to 20 MPa, were considerably lower than ultimate stress values in tension (around 100 MPa), but bone stresses were strongly affected by the implant geometry. The stress analysis showed that the 5 mm diameter Avinent implant had the best performance by reaching the lowest maximal principal stress values in the bone and inducing, at the same time, acceptable maximal principal stresses within the trabecular bone (Figure 5). The relative differences among implants were important; GMI implant (3.75 mm) almost tripled maximum principal stresses in the mandibular bone with respect to Avinent implant (5 mm) (see Figure 4, relative difference values). In order to compare the maximum principal stress levels in both cortical and trabecular periimplant regions, the cross-sections of mandibular bone are shown in Figure 5. Maximum principal stress concen-
trations appeared in the trabecular bone around the neck of the implants and at the screw threads, reaching values which varied from 1.8 MPa (Imp Avinent – 5 mm) to 5 MPa (Imp GMI – 3.75 mm). However, the Intri and GMI implants, both with a diameter of 3.75 mm, showed the highest stresses transferred to the mandibular bone with peak maximum principal stress values of almost 5 MPa within the trabecular bone which could cause long-term overloading risk. The average stress values at the bone in contact with the implant were very similar between Avinent and Lifecore implants (with 5 mm of diameter and similar geometrical characteristics). High averaged stress values were predicted for the GMI and Intri implants (with 3.75 mm of diameter). These implants presented clearly different geometrical characteristics with respect to Avinent and Lifecore implants. Stress distributions for the commercial dental implants modelled with a debonded bone – implant interface were very similar to the bonded bone – implant interface because of the high titanium stiffness. Nevertheless, maximum von Mises stresses on the implants with a debonded bone – implant interface were higher than for bonded interface. In contrast, the stress transfer at the debonded interface caused stress distributions completely different in the surrounding bone (Figure 6). Maximum principal stress levels were higher at the side where the load was applied. However, the results again showed the same tendencies: maximum principal stresses around the neck of the implants were reduced considerably with a large diameter.
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
268
M.A. Pe´rez et al.
Figure 4. Maximum principal stress (MPa) in mandibular bone for bonded bone– implant interface (B, buccal; D, distal; L, lingual and M, mesial).
Maximum principal stresses on the cross-sections of mandibular bone (Figure 7) were increased considerably with respect to a bonded interface (Figure 5). Higher maximum principal stresses were predicted for the Intri and GMI implants (3.75 mm) with peak stress values close to 6 MPa at the cross-section of mandibular bone
(Figure 7). The 5 mm Avinent implant showed the lowest maximum principal stress values at the cross-section of mandibular bone. Relative differences among implants were similar between both interface conditions (bonded or debonded without friction) (see Figures 4 and 6). Although higher averaged stress values at the bone in contact with
Figure 5. Maximum principal stress (MPa) at cross-sections of mandibular bone for bonded bone –implant interface (D, distal and M, mesial). Average value represented the average stress values for the bone in contact with the implant.
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
Computer Methods in Biomechanics and Biomedical Engineering
269
Figure 6. Maximum principal stress (MPa) in mandibular bone for debonded bone – implant interface (B, buccal; D, distal; L, lingual and M, Mesial).
the implant were predicted with a debonded without friction bone –implant interface than with a bonded bone – implant interface. Stabilisation of the implant during the healing time is a determinant factor for achieving success in immediate loading. Therefore, the tangential relative displacements between bone and implant are represented in Figure 8
considering the bone –implant interface as debonded. The analysis showed maximum micromovements for the Intri and GMI implants of 3.75 mm diameter with values of 9.23 and 10.31 mm, respectively. These higher micromovements were concentrated on smooth surfaces where the contacting area between bone and implant was reduced. Minimum micromotions were localised at the
Figure 7. Maximum principal stress (MPa) at cross-sections of mandibular bone for debonded bone – implant interface (D, distal and M, mesial). Average value represented the average stress values for the bone in contact with the implant.
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
270
M.A. Pe´rez et al.
Figure 8. Tangential relative motion (mm) at bone – implant interface for debonded interface (D, distal and M, mesial). Average value represented the average micromotion values for the bone in contact with the implant.
screw threads of the 5 mm diameter implants. No significant differences were found between the Intri and Lifecore implants (5 mm) regarding the lowest maximum relative micromotion values (Figure 8). Average micromotion values showed similar trends compared with the average stress values. In this case, Lifecore implant (5 mm) showed the lowest averaged relative micromotion at the bone – implant interface (Figure 8). This result could be directly related to its highest bone – implant interface surface (see Table 1).
4. Discussion The FE method has been increasingly adopted in the past few years to study the mechanical behaviour of biological structures (Prendergast 1997) and to predict the biomechanical performance of various dental implant designs as well as the effect of clinical factors on the success of implantation (Geng et al. 2001; van Staden et al. 2006). Dental implantology has high long-term success rates, although bone or implant failures are still unavoidable. A better understanding of load transfer mechanisms may help to further increase the success rates. Several FEAs have been previously performed to analyse the effect of implant design, diameter or bone – implant interface condition (Holmgren et al. 1998; van Oosterwyck et al. 1998; Chun et al. 2002; Himmlova´ et al. 2004). These involved several simplifications such as 2D FE models, implant as a plain cylinder and bone block instead of a real dental region. These limitations have been solved in the
present study where a 3D FEA was performed on four different commercial dental implants with two diameters and two boundary conditions. Stress transfer was reduced by increasing the diameter at both the implants and their surrounding bone as other studies of the literature have predicted (Holmgren et al. 1998; Himmlova´ et al. 2004; Ding et al. 2009; Figures 3– 7). However, implant design greatly influences the stress transfer distribution. The Intri implant has a very short threaded region (Figure 1) which does not help to distribute stress. The screw pitch of the GMI implant is higher than those of the Lifecore and Avinent implants (Figure 1), which reduces the contact area (Chun et al. 2002). This fact implies higher stresses at the GMI implant than those at the Lifecore and Avinent implants as it has been predicted herein. The best stress transfer distribution was predicted for the Avinent implant mainly due to its smaller width of thread and its threaded upper part of the implant-abutment which improves the stress state (Figure 1). The maximum principal stresses predicted for the different implants with the two bone – implant interface conditions were within a stress range of 1.4 –5 MPa (Figures 4 –7), which have been reported as necessary for maintaining existing cortical bone height in dental implants (Carter et al. 1996). Overloading of an implant may result in marginal bone resorption as has been demonstrated clinically (Quirynen et al. 1992). Most previous computational studies (Holmgren et al. 1998; Chun et al. 2002; Himmlova´ et al. 2004) considered the bone –implant interface as completely bonded, but we have also included the stress
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
Computer Methods in Biomechanics and Biomedical Engineering prediction for a debonded bone – implant interface (van Oosterwyck et al. 1998; Ding et al. 2009). Stresses were higher with debonded than with bonded bone – implant interface (van Oosterwyck et al. 1998). Although the results presented herein were in good agreement with previous observations (Holmgren et al. 1998; van Oosterwyck et al. 1998; Himmlova´ et al. 2004; Ding et al. 2009), several assumptions were made in the development of the present study. The components of the model were all assumed to be homogeneous and isotropic and possess linear elasticity. The properties of the materials modelled in this study, particularly the living tissues, are different. For instance, it is well documented that the cortical bone of the mandible is transversely isotropic and inhomogeneous. Therefore, the material properties assigned to each component will influence the stress distribution in the bone– implant system (Holmes and Loftus 1997; Eskitascioglu et al. 2004). Additionally, bone material properties were considered to remain constant, neglecting their evolutive behaviour. Herein, the lack of mechanical coupling between the machined coronal region of the implant and the bone, which avoids effective transfer of occlusal forces from the implant to the cortical bone, may also be related to the marginal bone resorption (Pilliar et al. 1991). Vaillancourt et al. (1995) observed that an equivalent stress equal to 1.6 MPa was determined to be sufficient to avoid bone loss due to disuse atrophy. Therefore, in the present study, some areas of crestal bone could experiment bone loss if bone evolution behaviour would have been considered for the four commercial dental implants. When applying FEA to dental implants, it is important to consider not only axial loads and horizontal forces but also a combined load (oblique occlusal force) because the latter represents more realistic occlusal directions, such as the one used in the present study (Holmgren et al. 1998; Sahin et al. 2002). The load values considered were average values. This underestimated the wide range of dispersion that has been observed experimentally and depended on the location of the implant in the dental arch (Mericske-Stern and Zarb 1996; Sahin et al. 2002). Average values were also considered for the material properties and geometrical characteristics. In fact, the endurance of implant fixation is directly related to the intrinsic stochastic evolution of these factors. In order to improve these results and conclusions, a probabilistic methodology would be more appropriate (Pe´rez et al. 2006). Loading application has also been simplified. Normally, loads are applied over the cusp of the crown whereas, herein, they have acted on the top section of the implant. Overloading has not been considered in the present study, although it has been reported to occur clinically (Quirynen et al. 1992). Overloading may be manifested by the application of repeated single loads,
271
which causes micro-fractures within the bone tissue; continuous application of low loads may also lead to failure, namely, stress or fatigue fracture (Rangert et al. 1997). Excessive dynamic loading may also decrease bone density around the neck of implants and lead to crater-like defects (Duyck et al. 2001). Additionally, no preload was considered (Alkan et al. 2004) because abutment and implant were treated as one component. Bone– implant interface conditions, as indicated, have a very strong effect on the bone-loading patterns around the implant. Not only the level of stresses and strains, but also the stress and strain distributions in bone are highly affected by the interface conditions (van Oosterwyck et al. 1998). Two extreme situations were modelled in the present study for the interface between bone and implant: bonded and debonded without friction. Both situations were completely idealised. Immediately after implant installation, implant and bone surfaces will not be congruent and, therefore, gaps may exist at the same locations. This is a limitation of the model. Both extreme situations could represent an osseointegrated interface if a correct functionality is achieved. Additionally, the bone – implant interface is a living surface that evolves from the latter to the former. Few studies have considered the evolution of the bone – implant interface properties. Bu¨cher et al. (2003) proposed a model in which the relative micromotions determined tissue differentiation (bone or fibrous tissue formation) at the bone –implant interface. Moreo et al. (2007) developed a phenomenological model able to simulate the mechanical effects of the osseointegration process taking into account both the different surface finishings and the influence of cyclic loading. Perez et al. (2008) applied the previous Moreo et al. (2007) model to predict the evolution of the resonance frequency analysis of the bone – implant interface in a dental implant. Their results were in quantitative agreement with some experimental measurements and they showed that the bone – implant interface was fully osseointegrated after several weeks (Perez et al. 2008). In the present study, the bone – implant interface was assumed as bonded in order to assess and compare the functional performance of different implants. The bone – implant interface was thus assumed to be debonded in order to analyse the relative displacements that may interfere with the osseointegration process (Figure 8). Implants subjected to relative displacements higher than 50 mm were found to be covered by an interfacial fibrous membrane; therefore, no osseointegration (no clinical functionality) was achieved (Kienapfel et al. 1999). Nevertheless, the relative displacements predicted at the interface (Figure 8) were within the limits required to achieve a good osseointegration which considers micromotions below 50 mm as do not have an influence on the osteogenesis and bone remodelling (Kienapfel et al. 1999). Large relative motion would have been predicted if a lower elastic modulus of trabecular bone would have been assumed
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
272
M.A. Pe´rez et al.
(Misch et al. 1999), additionally full bone contact was considered at the bone – implant interface neglecting the incongruent surfaces due to implantation procedure. Only part of the mandible has been modelled; this fact may influence local displacements, stresses and strains in the peri-implant region underestimating the relative motions computed. The evolutive behaviour of the bone – implant interface should have been considered in order to analyse the effect of biological events in the general behaviour of the bone –implant system. Clinical results may help to guarantee this kind of FEA. Clinical results have been only reported for Lifecore implants among the four commercial dental implants considered in the present study. Clinical performance of Lifecore implants was demonstrated by Kallus et al. (2009), where they compared survival rates and marginal bone resorption of the Lifecore system with Nobel Biocare one. No significant differences were found between the two implant systems regarding survival rates. Although significantly more bone loss could be demonstrated for the Nobel Biocare implants after 5 years than for the Lifecore system. In conclusion, the proposed 3D FE simulation approach can be considered as an accurate enough tool for performing stress evaluation of different commercial dental implants. It may allow a quantitative and qualitative comparison between different commercial dental implants, and different design factors (diameter, length, screw type, thread pitch and thread depth). It may be also applied to study stress and relative motion, although we should take into account the previously cited limitations when trying to extend the conclusions obtained to the actual behaviour of available commercial dental implants.
Acknowledgements The authors gratefully acknowledge the research support of the Spanish Ministry of Science and Technology through the Research Project DPI2008-02335, the Carlos III Health Institute (CIBER-BBN) and the Aragon Institute of Engineering Research (I3A) for their graduate research fellowship program. Special acknowledgments are be made to Avinent Implant system company, Ilerimplant, Lifecore and Intri. All the authors of this paper disclose no conflict of interest with previous companies.
References Adell R, Erikkson B, Lekholm U, Branemark PI, Jemt T. 1990. A long-term follow up study of osseointegrated implants in the treatment of totally edentulous jaws. Int J Oral Maxillofac Implants. 5:347 – 359. Albrektsson T, Wennerberg A. 2005. The impact of oral implants – past and future, 1966– 2042. J Can Dent Assoc (Tor). 71:327. Alkan I, Sertgo¨z A, Ekici B. 2004. Influence of occlusal forces on stress distribution in preload dental implant screws. J Prosthet Dent. 91:319 –325.
Asmussen E, Peutzfeldt A, Sahafi A. 2005. Finite element analysis of stresses in endodontically treated, dowel-restores teeth. J Prosthet Dent. 94:321 – 329. Baggi L, Cappelloni I, Maceri F, Vairo G. 2008. Stress-based performance evaluation of osseointegrated dental implants by finite-element simulation. Simul Model Pract Theory. 16:971 – 987. Bidez MW, Misch CE. 1992. Force transfer in implant dentistry: basic concepts and principles. J Oral Implantol. 18:264– 274. Bozkaya D, Muftu S, Muftu A. 2004. Evaluation of load transfer characteristics of five different implants in compact bone at different load levels by finite element analysis. J Prosthet Dent. 92:523 – 530. Branemark PI, Breine U, Adell R, Hansson BO, Lindstrom J, Oholsson A. 1969. Intra-osseous anchorage of dental prostheses. I. Experimental studies. Scand J Plast Reconstr Surg. 3:81 – 100. Brunski JB. 1992. Biomechanical factors affecting the bone – dental implant interface. Clin Mater. 10:153 –201. Bu¨cher P, Pioletti DP, Rakotomanana LR. 2003. Biphasic constitutive laws for biological interface evolution. Biomech Model Mechanobiol. 1:239– 249. Carter DR, van der Meulen MC, Beaupre GS. 1996. Mechanical factors in bone growth and development. Bone. 18(Suppl):S5– 10. Chun HJ, Cheong SY, Han JH, Heo JP, Chung JP, Rhyu IC, Choi YC, Baik HK, Ku Y, Kim MH. 2002. Evaluation of design parameters of osseointegrated dental implants using finite element analysis. J Oral Rehabil. 29:565 – 574. Chun HJ, Park DN, Han CH, Heo SJ, Heo MS, Koak JY. 2005. Stress distributions in maxillary bone surrounding overdenture implants with different overdenture attachments. J Oral Rehabil. 32:193– 205. Chun HJ, Shin HS, Han CH, Lee SH. 2006. Influence of implant abutment type on stress distribution in bone under various loading conditions using finite element analysis. Int J Oral Maxillofac Implants. 21:195 – 202. Cook RD, Malkus DS, Plesha ME. 2001. Concepts and applications of finite element analysis. 4th ed. New York: Wiley. p. 310– 314. Ding X, Liao SH, Zuh XH, Zhang XH, Zhang L. 2009. Effect of diameter and length on stress distribution of the alveolar crest around immediate loading implants. Clin Imp Dent Rel Res. 11(4):279– 287. Duyck J, Ronold HJ, Van Oosterwyck H, Naert I, Vander Sloten J, Ellingsen JE. 2001. The influence of static and dynamic loading on marginal bone reactions around osseointegrated implants: an animal experimental study. Clin Oral Implants Res. 12:207 – 218. Eskitascioglu G, Usumez A, Sevimay M, Soykan E, Unsal E. 2004. The influence of occlusal loading location on stresses transferred to implant-supported prostheses and supporting bone: a three-dimensional finite element study. J Prosthet Dent. 91:144 – 150. Geng JP, Tan KBC, Liu GR. 2001. Application of finite element analysis in implant dentistry: a review of the literature. J Prosthet Dent. 85:585– 598. Himmlova´ L, Dosta´lova´ T, Ka´covsky´ A, Konvicˇkova´ S. 2004. Influence of implant length and diameter on stress distribution: a finite element analysis. J Prosthet Dent. 91:20 –25. Holmes DC, Loftus JT. 1997. Influence of bone quality on stress distribution for endosseous implants. J Oral Implantol. 23:104 – 111.
Downloaded by [UGR-BTCA Gral Universitaria] at 06:43 21 March 2012
Computer Methods in Biomechanics and Biomedical Engineering Holmgren EP, Seckinger RJ, Kilgren LM, Mante F. 1998. Evaluating parameters of osseointegrated dental implants using finite element analysis – a two-dimensional comparative study examining the effects of implant diameter, implant shape, and load direction. J Oral Implantol. 24:80 – 88. Jaffin RA, Kumar A, Berman CL. 2004. Immediate loading of dental implants in completely edentulous maxilla: a clinical report. Int J Oral Maxillofac Implants. 19:721 – 730. Kallus T, Bessing C, Homsi G, Eklund I. 2009. Five-year evaluation of Lifecore Restore implants: a retrospective comparison with Nobel Biocare MK II implants. Clin Implant Dent Relat Res. 11(3):167– 177. Kienapfel H, Sprey C, Wilke A, Griss P. 1999. Implant fixation by bone in-growth. J Arthroplasty. 14:355 – 368. Kitagawa T, Tanimoto Y, Nemoto K, Aida M. 2005. Influence of cortical bone quality on stress distribution in bone around dental implant. Dent Mater J. 24:219 –224. Maceri F, Martignoni M, Vairo G. 2007. Mechanical behaviour of endodontic restorations with multiple prefabricated posts: a finite element approach. J Biomech. 40:2386– 2398. Martin RB, Burr DB, Sharkey NA. 1998. Skeletal tissue mechanics. New York: Springer-Verlag. p. 127– 224. Menicucci G, Mossolov A, Mozzati M, Lorenzetti M, Preti G. 2002. Tooth-implant connection: some biomechanical aspects based on finite element analyses. Clin Oral Implants Res. 13:334 – 341. Mericske-Stern R, Zarb GA. 1996. In vivo measurements of some functional aspects with mandibular fixed prostheses supported by implants. Clin Oral Implants Res. 7:153 –161. Misch CE, Qu Z, Bidez MW. 1999. Mechanical properties of trabecular bone in the human mandible: implications for dental implant treatment planning and surgical placement. J Oral Maxillofac Surg. 57:700 – 706. Miura H, Hasaqawa S, Okada D, Ishihara H. 1998. The measurement of physiological tooth displacement in function. J Med Dent Sci. 45:103 – 115. Moreo P, Pe´rez MA, Garcı´a-Aznar JM, Doblare´ M. 2007. Modeling the mechanical behaviour of living bony interfaces. Comput Methods Appl Mech Eng. 196:3300 – 3314. Moreo P, Garcı´a-Aznar JM, Doblare´ M. 2009. Bone in growth on the surface of endosseous implants. Part 1: mathematical model. J Theor Biol. 260(1):1– 12. Ostman PO, Hellman M, Sennerby L. 2008. Immediate occlusal loading of implants in the partially edentate mandible: a prospective 1-year radiographic and 4-year clinical study. Int J Oral Maxillofac Implants. 23:315– 322. Pe´rez del Palomar A, Arruga A, Cegon˜ino J, Doblare´ M. 2005. A finite element comparison between the mechanical behaviour of rigid and resilent oral implants with respect to
273
immediate loading. Comput Methods Biomech Biomed Engin. 8:45– 57. Pe´rez MA, Prendergast PJ. 2007. Random-walk models of cell dispersal included in mechanobiological simulations of tissue differentiation. J Biomech. 40(10):2244– 2253. Pe´rez MA, Grasa J, Garcı´a-Aznar JM, Bea JA, Doblare M. 2006. Probabilistic analysis of the influence of the bonding degree of the stem – cement interface in the performance of cemented hip prostheses. J Biomech. 39:1859 – 1872. Perez MA, Moreo P, Garcı´a-Aznar JM, Doblare´ M. 2008. Computational simulation of dental implant osseointegration through resonant frequency analysis. J Biomech. 41:316–325. Petrie CS, Williams JL. 2005. Comparative evaluation of implant designs: influence of diameter, length, and taper on strains in the alveolar crest. A three-dimensional finite-element analysis. Clin Oral Implants Res. 16:486 – 494. Pikner SS. 2008. Radiographic follow-up analysis of Bra˚nemark dental implants. Swed Dent J Suppl. 194:5 – 69. Pilliar RM, Deporter DA, Watson PA, Valiquette N. 1991. Dental implant design-effect on bone remodelling. J Biomed Mater Res. 25:467 – 483. Prendergast PJ. 1997. Finite element models in tissue mechanics and orthopaedic implant design. Clin Biomech. (Bristol, Avon) 12:343 – 366. Quirynen M, Naert I, van Steenberghe D. 1992. Fixture design and overload influence marginal bone loss and fixture success in the Branemark system. Clin Oral Implants Res. 3:104 –111. Rangert BR, Sullivan RM, Jemt TM. 1997. Load factor control for implants in the osterior partially edentulous segment. Int J Oral Maxillofac Implants. 12:360 – 370. Sahin S, Cehreli MC, Yalcin E. 2002. The influence of functional forces on the biomechanics of implant-supported prostheses – a review. J Dent. 30:271– 282. Sevimay M, Turhan F, Kilicarslan MA, Eskitascioglu G. 2005. Three-dimensional finite element analysis of the effect of different bone quality on stress distribution in an implantsupported crown. J Prosthet Dent. 93:227– 234. Vaillancourt H, Pilliar RM, McCammond D. 1995. Finite element analysis of crestal bone loss around porous-coated dental implants. J Appl Biomater. 6(4):267 – 282. van Oosterwyck H, Duyck J, Vander Sloten J, Van der Perre G, de Cooman M, Lievens S, Puers R, Naert I. 1998. The influence of bone mechanical properties and implant fixation upon bone healing loading around oral implants. Clin Oral Implants Res. 9:407– 418. van Staden RC, Guan H, Loo YC. 2006. Application of the finite element method in dental implant research. Comput Methods Biomech Biomed Engin. 9:257 – 270.
