The effect of screw preload and framework material on the success of cementable fixed partial prostheses: A finite element study

The effect of screw preload and framework material on the success of cementable fixed partial prostheses: A finite element study

Accepted Manuscript Title: The Effect of Screw Preload and Framework Material on the Success of Cementable Fixed Partial Prostheses: A Finite Element ...

598KB Sizes 0 Downloads 27 Views

Accepted Manuscript Title: The Effect of Screw Preload and Framework Material on the Success of Cementable Fixed Partial Prostheses: A Finite Element Study Author: Istabrak Hasan Ludger Keilig Christoph Bourauel Walter L¨uckerath PII: DOI: Reference:

S0940-9602(14)00035-1 http://dx.doi.org/doi:10.1016/j.aanat.2014.03.005 AANAT 50862

To appear in: Received date: Revised date: Accepted date:

31-10-2013 7-3-2014 31-3-2014

Please cite this article as: Hasan, I., Keilig, L., Bourauel, C., L¨uckerath, W.,The Effect of Screw Preload and Framework Material on the Success of Cementable Fixed Partial Prostheses: A Finite Element Study, Annals of Anatomy (2014), http://dx.doi.org/10.1016/j.aanat.2014.03.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The Effect of Screw Preload and Framework Material on the Success of Cementable Fixed Partial Prostheses: A Finite Element Study

Endowed Chair of Oral Technology, Rheinische Friedrich-Wilhelms University,

cr

1

ip t

Istabrak Hasan1, Ludger Keilig1,2, Christoph Bourauel1, Walter Lückerath2

2

us

Welschnonnenstr. 17, 53111 Bonn, Germany

Department of Prosthetic Dentistry, Preclinical Education and Materials Science,

an

Dental School, Rheinische Friedrich-Wilhelms University, Welschnonnenstr. 17,

M

53111 Bonn, Germany

te

d

Running title: Cementable implant-supported fixed prosthesis

Ac ce p

Correspondence to: Dr. rer. nat. Istabrak Hasan, Endowed Chair of Oral Technology, Department of Prosthetic Dentistry, Preclinical Education and Materials Science,

Dental

School,

Rheinische

Friedrich-Wilhelms

University,

Welschnonnenstr. 17, 53111 Bonn, Germany. Tel: +49 228 287 22388. E-mail: [email protected].

Words in abstract: 246 Words in the text: 3,764

1

Page 1 of 41

Abstract The rigidity of framework materials and overload of the implant system directly affect the final transferred load of the bone around implants. The aim of the present study

ip t

has been to analyse the influence of framework materials on the transferred load to the implant system and the surrounding bone.

cr

A finite element model of a long-span cementable implant-supported fixed prosthesis

us

was created with two coping layers (gold and hybrid composite) to optimise the fitting of the prosthesis to the abutments. Three framework materials were analysed:

an

Titanium, gold alloy, and zirconia. The connection screws were first preloaded with 200 N. The framework was then loaded with 500 N vertically and at 30° to the

M

framework long axis. Two loading conditions were considered: At the mesial and distal boundaries of the framework and at the centre of the framework.

d

The stresses and strains within the framework materials and bone bed around the

te

supporting implants were analysed. The region and angle of load applications showed an obvious effect on the values of the stresses and strains within the

Ac ce p

framework itself and, consequently, their distribution in the implant system and surrounding bone. A correlation of the framework material and stresses of the coping materials was observed as well. The gold framework showed acceptable values of stress within the cortical bone (92 MPa and 89 MPa with 30° loading at two points and at the centre, respectively) in comparison to titanium (92 MPa and 113 MPa) and zirconia (88 MPa and 115 MPa).

Key words: preload, framework, misfit, stress, strains.

2

Page 2 of 41

Introduction The final goal of implant-prosthetic treatment is an aesthetic and most of all functional restoration, preventing any implant component from possible collapse

ip t

(Albrektsson, 1988; Quirynen et al., 2002; Oh et al., 2002). Implant failure may depend on two distinct types of factors, biological and mechanical. Biological causes

cr

are essentially peri-implantitis, affecting the soft and hard tissues surrounding dental implants, while mechanical causes involve implant-prosthetic components at large.

us

Mechanical complications are: Implant fracture, abutment fracture, screw loosening,

an

and loss, as well as over-structure (ceramic and /or metal) fracture (Albrektsson, 1988; Quirynen et al., 2002).

M

The clamping force from the retention screw provides a stable connection between the abutment and the dental implant (McGlumpy et al., 1998; Sakaguchi et al.,

d

1994). Preload is the term given to the tension generated in the screw upon

te

tightening and is a direct determinant of clamping force. This preload is generated by rotational torque that elongates the screw within the material yield strength (for an

Ac ce p

abutment retention screw, generally, from 25 to 35 Ncm, Griffith, 1987). The elasticity of the material used in screw manufacture is important in the development and maintenance of preload (Kim et al., 2005). Optimal preload for a screw is achieved when the screw is elongated, but not to a point at which the yield strength is exceeded. Implant abutment screws are most often made of titanium alloys or gold alloys (Byrne et al., 2006). Excessive tightening torque can provide higher preload for a more stable implantabutment connection joint. However, this method may introduce additional rotational and shearing forces into the implant system, particularly when they are placed in

3

Page 3 of 41

soft-quality bone, consequently interfering with the osseointegration process of the bone around the implant (Park et al., 2010). Implant-abutment misfit is known to increase mechanical stress on connection

ip t

structures and surrounding bone tissue. This condition may induce screw preload loss or fracture, and cause biological issues due to bacterial penetration within a

cr

possible fixture-abutment gap (Albrektsson, 1988; Bickford; 1981; Coelho et al., 2008; Jorneus et al., 1992; McGlumpy et al., 1998; Meleo et al., 2012; Oh et al.,

us

2002; Patterson and Johns, 1992; Quirynen et al., 2002).

an

Frameworks should be biocompatible, have excellent physical properties in terms of strength, and fit accurately to implants and abutments (Mericske-Stern, 2008). A

M

passive fit of implant-supported prostheses is considered a prerequisite to the prevention of mechanical complications (Greven et al., 2007), and therefore

d

prosthetic success. Because implants lack the stress release associated with a

te

periodontal ligament, impact loading to restorative materials and the crestal bone remains potentially more damaging with implant-supported restorations (Curtis et al.,

Ac ce p

2000). It is therefore believed that dental implants may be more prone to occlusal overloading, which is often regarded as one of the potential causes of periimplant bone loss and failure of implant supported prostheses (Kim et al., 2005). Two basic methods are currently used in the fabrication of implant frameworks: The conventional lost wax/casting technique (Puri, 2005) and CAD/CAM milling procedures where frameworks are milled from solid blanks of titanium, titanium alloy, or ceramic materials such as zirconia (Jemt and Petersson, 1999). The benefits of the lost wax/casting technique include the ability to create optimal aesthetics due to the proven technology associated with porcelain fused to metal (Segal, 2001), high biocompatibility with gold alloys (Craig and Hanks, 1990; Kansu and Aydin; 1996),

4

Page 4 of 41

and the ability of most commercial dental laboratories to fabricate implant frameworks with this proven technology (Maló et al., 2012). Zirconia has been used extensively for milled cementable frameworks that have a lower fabrication cost than

ip t

cast frameworks. The use of a device to verify the accuracy of implant positions on the definitive cast is recommended because sectioning and soldering of milled

cr

zirconia frameworks is not possible (Alhashim et al., 2012).

The load transmission to the prosthesis, implants, and bone depends on several

us

factors, as the number and location of the implants (Karl et al., 2007; Ogawa et al.,

an

2010; Sagat et al., 2010), inclination of the implants (Bevilacqua et al., 2011; Markarian et al., 2007), stiffness of the metal framework (Abreu et al., 2010)

M

prosthesis marginal fit (Markarian et al., 2007; Carr et al., 1996; Winter et al., 2010), prosthesis material (Ogawa et al., 2010), extension of the prosthesis base and

d

attachment systems (Sadowsky and Caputo, 2000), and occlusion pattern (Greco et

te

al., 2009; de Torres et al., 2011).

Hence, it was the aim of the present study to analyse the influence of framework

Ac ce p

materials on the transferred load to the implant system and the surrounding bone by assuming an optimal passive fit of the framework to the abutment using a numerical model. The effect of retention screw preloading was considered on the distribution of the force within the implant system.

5

Page 5 of 41

Materials and Methods A three-dimensional finite element (FE) model of a cementable fixed partial prosthesis was constructed at the lower left posterior region. A long-span bridge from

ip t

the canine to second premolar was considered to analyse an extreme loading condition of the prosthesis. Two implants were used to support the prosthesis in its

cr

most mesial and distal boundaries (Figure 1). Straumann Bone Level Implants (4.1x11 mm) were used in this study. After scanning the selected implant system and

us

the corresponding abutments with the aid of µCT device (SkyScan 1174, SKYSCAN,

software ADOR3D (Rahimi et al., 2005).

an

Belgium), the geometry of the implants was reconstructed using the self developed

M

The implants were inserted in the lower left posterior segment of the mandible and an osseointegrated condition was considered. The bone segment was taken from an

d

individual mandibular model that was created from CT-data of an anonymous

te

patient. The thickness of the cortical layer in this region was about 1.0 mm. Individual abutments were later placed in their optimal fitting position of the implants

Ac ce p

by considering the preloading of the connecting screw with a moment of 35 Ncm. The preload was simulated in the model by using a special option in the FE package Marc/Mentat 2010 (MSC.Software, Santa Ana, CA-USA) that allows for loading the connected screw up to the predefined load and in which the optimal contact to the abutment is reached. The predefined load was 200 N which corresponds to 35 Ncm. Since obtaining the optimal fitting of the prosthesis on the abutments is an important problem clinically and decisive to the success of the overall treatment, one approach to minimise the misfit is to consider several coping layers of the inner surface of the framework in connection to the abutment. Accordingly, the framework of the numerical model was developed with two copings at the fitting position to the

6

Page 6 of 41

abutments, namely, 0.30 mm pure gold coping and 0.25 mm of hybrid composite (Figure 1b). Three materials of the framework were tested in this study; i.e., titanium alloy (Grade

ip t

5), zirconia, and gold alloy type III. Since veneer materials were not in the focus of the study, the FE model included only the framework without veneer. The

cr

components of the model and their material properties are summarized in Table 1. The FE package Marc/Mentat 2010 (MSC.Software, Santa Ana, CA-USA) was used

us

to create the numerical model. The element type used was a 4-noded tetrahedral

an

element. The final models had a total number of 601,959 elements and 131,862 nodes. An adaptive meshing with a coarsening factor of 1.5 was used to create the

M

volume meshing. The length of element edge thereby was adapted according the size and complexity of the individual geometries.

d

The loading of the model was realized in two steps: Firstly, the retention screws were

te

preloaded with a force of 200 N (35 Ncm, correspondingly) to obtain the resulting stresses of the system (abutment, screw, and implant) before applying the main load

Ac ce p

on the prosthesis. Secondly, the framework was loaded with a total force of 500 N. Two loading directions were studied: a) Perpendicular to the long axis of the framework (Figure 2, black lines) and b) 30° from its long axis in bucco-lingual direction (Figure 2, red lines). For each loading direction, two loading positions were analysed: The whole force was applied at the canine and second molar regions of the framework (250 N each, Figure 2, 1a and 1b) or the 500 N was applied at the centre of the framework (Figure 2, 2a and 2b). An overview of the loading conditions of the model is illustrated in Figure 2.

7

Page 7 of 41

Results The stresses generated in the cortical bone and strains in the cancellous bone with vertical and oblique loading of the framework are presented below. Moreover, the

ip t

stresses and strains within the three different framework materials are analysed and compared to the yield strength of the tested materials. Preloading the retention

cr

screws with 200 N (35 Ncm) before applying loading of the framework caused a stress of 16 MPa within the cortical bone. This value was omitted from the finally

us

generated stresses of the cortical bone and was not considered in the evaluation of

an

the final stresses within the cortical bone.

M

Vertical Loading of the Framework

For the framework itself, the maximum stress of the different framework materials

d

was around 600 MPa, when the prosthesis was loaded with 500 N at two points.

te

There was no noticeable difference in the distribution of the stress for the three framework materials (Figure 3a, left side Figures). These observed stresses were

Ac ce p

much lower than the yield strength of titanium grade 5 and zirconia. However, the maximum stress with the gold alloy exceeded the yield strength of the gold alloy type III (Table 2, Figure 3b). By applying the load at two points of the framework, the highest strain was observed with the gold framework (Table 3; Figure 4, left side Figures). There was no noticeable difference in the distribution of the strain for the three framework materials (Figure 4a, left side Figures). By applying the whole load on the centre of the framework, a critical increase in the stress within the framework values was observed (Figure 3, left side Figures). For the zirconia framework, the stress was 1,203 MPa which is slightly above the yield strength of 1,000 MPa. The stresses for the titanium and gold frameworks were

8

Page 8 of 41

significantly higher than the yield strength (1,795 MPa and 1,757 MPa, respectively). The highest strain for this loading case was observed with the gold framework as well (Figure 4b, left). The distribution of the strain was similar for the gold and

ip t

titanium frameworks (Figure 4a, left side Figures). The obtained stresses for the first coping layer (pure gold) were above the yield

cr

strength of 112 MPa, in particular when the load was applied on the centre of the framework, where the stress within the pure gold coping reached 474 MPa with the

us

gold framework and 440 MPa with titanium framework. For the composite coping

an

layer, the obtained stresses were below the yield strength of 450 MPa, except for the titanium and gold frameworks, the stresses were obviously above the yield strength

M

of hybrid composite (617 MPa and 646 MPa, respectively) when the load was applied vertically on the centre (Table 2).

d

The stresses in the implant system (implant, abutment, and retention screw) were in

te

general below the yield strength of titanium grade 5. However, the stress values were increased by changing the position of the load application from the two

Ac ce p

framework borders to the centre, where the highest stress was obtained with the titanium framework (880 MPa; Figure 3, left side Figures). The highest stresses for all models were concentrated on the lingual side of implants’ shoulder and the lower thread region of the connection screw (Figure 3a, left side Figures). Titanium framework showed the lowest stress in the cortical bone (69 MPa) in comparison to zirconia (76 MPa) and gold (87 MPa), when the vertical load was applied on two ends of the framework. However, the stress distributions within the cortical bone were very similar for the three framework materials (Figure 5, left side Figures). The strain values of the cancellous bone for the three materials were 3,409 µε, 4,000 µε, and 3,000 µε for titanium, gold, and zirconia frameworks, respectively

9

Page 9 of 41

(Figure 6). An obvious reduction in the strain of the cancellous bone was observed with the zirconia framework (Figure 6a, left side Figures). By changing the region of vertical load application to the centre, the stresses of the

ip t

cortical bone showed a reduction in their values, namely, 44 MPa for gold, while for zirconia and titanium there was an increase in the stresses of the cortical load (89

cr

MPa and 78 MPa, respectively) and wider distribution in comparison to the application of the vertical loads on two points as can be seen in Figure 5 (left side

us

Figures of Figure 5b). However, the strains within the cancellous bone showed an

an

increase in their values for the three framework materials by applying the total load vertically on the centre of the framework. The strain values were 5,500 µε, 6,000 µε,

M

and 3,429 µε for titanium, gold, and zirconia frameworks, respectively (Table 3, Figure 6b). Zirconia framework showed the lowest strain in the cancellous bone by

te

d

applying the load to the centre (Figure 6a).

Oblique Loading of the Framework

Ac ce p

By applying the load at an angle of 30° in bucco-lingual direction, there was a threefold increase in the obtained stress values in comparison to those observed by loading the framework vertically. By applying 500 N at two points, the maximum stresses were 1,441 MPa, 1,446 MPa, and 1413 MPa for titanium, gold, and zirconia frameworks, respectively and 2,418 MPa, 2,376 MPa, and 2,488 MPa by applying the 500 N at the centre of the framework (Table 2; Figure 3, right side Figures). The strains showed a credible increase in their values in comparison to the vertical loading of the framework (Table 3, Figure 4). Gold frameworks showed the highest strain values, namely, 0.015 and 0.028 for loading at two points and at the centre, respectively.

10

Page 10 of 41

The obtained stresses for the first coping layer (pure gold) were above the yield strength of 112 MPa, in particular when the load was applied on the two ends of the framework, where the stress within the pure gold coping reached 430 MPa with the

ip t

gold framework and 403 MPa with the titanium framework. Concerning the composite coping layer, the obtained stresses were close to the yield

cr

strength of 450 MPa. However, by applying the total load at the centre, the stress of composite coping was 512 MPa with the zirconia framework followed by 460 MPa for

us

the titanium framework (Table 2).

an

By loading the framework on the centre, the stresses within the implant system (implant, abutment, and retention screw) were very close to the yield strength of

M

titanium grade 5, it was even exceeded with the zirconia framework (970 MPa). The highest stresses were concentrated at the implant-abutment, implant-screw

d

interfaces, and the lower thread part of the retention screw as well. (Figure 3a, right

te

side Figures).

Within the cortical bone, the stresses were around 92 MPa for the three framework

Ac ce p

materials by applying the oblique load at two points, while a difference in the values of the stresses within the cortical bone was observed by applying the load on the centre: The lowest stress was determined with the gold framework (89 MPa), while cortical bone stress was 113 MPa for titanium and 115 MPa for zirconia (Table 2; Figure 5, right side Figures).

The strain values in the cancellous bone were 8,500 µε, 7,200 µε, and 7,500 µε for titanium, gold, and zirconia frameworks, respectively, by applying the oblique load at two points, while by applying the load at the centre, the strains were 7,092 µε, 6,000 µε, and 8,000 µε for titanium, gold, and zirconia frameworks, respectively (Table 2; Figure 6, right side Figures).

11

Page 11 of 41

The application of the load at an angle of 30° on the framework instead of vertically, showed a critical change in the distribution of the strains within the cancellous bone

Ac ce p

te

d

M

an

us

cr

ip t

(Figure 6a, compare left side for vertical load and right side for oblique load).

12

Page 12 of 41

Discussion Implant-supported fixed prostheses are widely used as an alternative treatment for edentulous patients. Cementable fixed prostheses with temporary cement material

ip t

have the advantage of easy removal of the prostheses from the supporting abutments and, for this reason, offer greater flexibility in repair of prosthetic

cr

complications as compared to screw-retained prostheses, that are mostly combined with screw loosening or even fracture after a certain period of functional loading of

us

the prosthesis.

an

It is important to control implant occlusion within physiological limits and thus to provide optimal implant load to ensure long-term implant success, but currently there

M

is no evidence-based implant specific concept of occlusion (Kim et al., 2005; Maló et al., 2012). A mathematical model by means of finite element analysis is superior to in

d

vivo tests in aspects of repeatability and controllability (Simsek et al., 2006), in

te

particular, for analysing the influence of the magnitude and direction of the occlusal loads on the stress distribution from the superstructure into the abutment, implant

Ac ce p

and thereby to the surrounding alveolar bone. This is not only affected by the occlusal loads, but by the insertion angle of the implant to the final restoration as well, where the actual load is applied. For this, in our study, we aimed to have realistic bone geometry as an implant bed based on CT-date of an edentulous patient. This had the advantage that the final position of the implants was dependent on the available contour of the alveolar bone which is clinically relevant. In our model, the final position of the implant at the second molar region was not parallel to the long axis of the restoration. The goal of the present study was to investigate the load distribution within the framework, implant system and the surrounding bone with the commonly used materials of the framework,

13

Page 13 of 41

namely, titanium, gold alloy, and zirconia with the consideration of the effect of insertion angle of the supporting implants and the preload of retention screws on the resulting stresses and strains of the different components of the prosthesis, implant

ip t

system and alveolar bone. Because of the angulated position of the implant in relation to the framework, vertical

cr

loading of the framework caused concentration of the stresses on the lingual side of the implant shoulder and the thread region of the retention screw, in particular when

us

the framework was loaded at the centre. The critical value of the stress was

an

determined with the titanium framework (880 MPa). Applying the load obliquely on the framework resulted in more critical concentration of the stress within the implant

M

system and increased the risk of an overload and fatigue failure of the implant system material. The most critical value with oblique force was registered with the

d

zirconia framework.

te

The absolute values for preload within the implant, prosthesis, abutment, and screws vary considerably among studies, and this may be due to differences in how preload

Ac ce p

is measured (Hagiwara and Ohashi, 1994). Preload has been calculated in several experiments from rotational angle (Martin et al., 2001), from compression in the implant assembly (Cantwell and Hobkirk, 2004; Tan and Nicholls, 2002), or from screw elongation (Byrne et al., 2006; Haack et al., 1995). Screw-tightening torque may primarily elongate screws rather than deform the framework because of the interface on the initial contact points between the framework and implants. In particular, screw-tightening torque may be considered to generate preload stresses on the screws rather than on the framework. When functional loads are superimposed on preload stresses, a settling down of the framework with additional stresses can result over time. These stresses can contribute to loosening or fracture

14

Page 14 of 41

of the implant and prosthetic components of the restorative complex (Choi et al., 2009). Since the fixation of the abutment on the implant by the retention screw results in loading of the implant system to a certain degree, in our FE models, a

ip t

preload of 200 N (corresponding to 35 Ncm of insertion torque) of the retention screws was simulated by means of elongation prior to application of the load on the

cr

framework.

In the clinical report of Rojas-Vizcaya (Rojas-Vizcaya, 2011), it was mentioned that

us

in studies of hybrid prostheses using frameworks of various materials, several

an

different complications arose. In 1999, Bergendal and Palmqvist compared titanium frameworks and gold alloys over 5 years, and reported slightly higher fracture rates

M

of titanium frameworks than gold alloy frameworks and more fractures of artificial teeth in the titanium frameworks. Most fractures were related to the welding joints at

d

the distal abutments. In 2009, Örtorp and Jemt conducted a comparative follow-up

te

study on a supervised period of 15 years, in which laser-welded titanium frameworks were compared with gold alloy frameworks. Fractures in the titanium framework were

Ac ce p

detected in 15.5% of the patients. More fractures were detected in titanium frameworks than in gold alloy frameworks (Örtorp and Jemt, 2009). The stress results in this study showed that the risk of framework fracture for titanium and gold alloy is high in comparison to zirconia, in particular under oblique load at the centre. One of the solutions to overcome the misfit of frameworks and improve their passivity is to apply copings of other materials than used for the framework at the interface with the abutments. This could indeed improve the passive fitting of the framework but it could cause material failure of the coping and/or change in the load transfer into the surrounding bone because of the differences in the stiffness and deformation between framework and coping materials. In the present study, it was

15

Page 15 of 41

observed that the stresses of the gold coping were clearly higher than the yield strength of the pure gold with the three different frameworks. In addition to the degree of fit, the elastic modulus of framework materials can

ip t

influence the peri-implant stress and strain. In a FE analysis, Rubo and Souza46 found that the stiffer the implant framework, the more uniform the stress was

cr

distributed within the peri-implant bone (Rubo and Souza, 2008). In contrast, a series of strain gauge analysis studies found that the stiffer Co-Cr framework resulted in

us

more deformation of the abutments when compared to a less stiff alloy (silver

an

palladium) under loading conditions (Jacques et al., 2009; Suedam et al., 2009). The differences between the FE studies and strain gauge analysis studies can be

M

explained by the fact that FE studies had an assumed optimal fit, which eliminated the possibility of strain transmission as a result of inserting a miss fitting prosthesis.

d

Stiffer frameworks are more resistant to bending as the retaining screws are

te

tightened, leading to more strain transmission to implant components or the surrounding bone (Abduo et al., 2011).

Ac ce p

The strain results of the cancellous bone showed that the magnitude of the strain is more affected by the loading condition of the framework rather than the type of the framework material itself. By applying the load vertically on the two ends of the framework or on the centre, the less stiff material; i.e. gold alloy, showed the highest strain of 4,000 µε and 6,000 µε, respectively, in comparison to titanium and zirconia. In contrast, changing the loading direction from vertical to oblique loading of 30° caused higher strains of the cancellous bone with the stiffer framework materials; i.e. zirconia and titanium, than with gold alloy.

16

Page 16 of 41

Conclusions 1.

Increase in the insertion angle of the implants to the connected final restoration causes an increase of the risk of fatigue failure of the supporting

2.

ip t

implants and/or the retention screw with zirconia framework. Combining zirconia framework with pure gold coping can reduce the stress

The magnitude of the stresses and strains of the alveolar bone is more related

us

to the occlusion and loading conditions of the framework than the type of the

te

d

M

an

framework material.

Ac ce p

3.

cr

within the coping material.

17

Page 17 of 41

References Abduo, J., Bennani, V., Lyons, K., Waddell, N., Swain, M., 2011. A novel in vitro approach to assess the fit of implant frameworks. Clin. Oral Implants Res. 22, 658-

ip t

663.

cr

Abreu, R.T., Spazzin, A.O., Noritomi, P.Y., Consani, R.L.X., Mesquita, M.F., 2010. Influence of material of overdenture-retaining bar with vertical misfit on three-

an

us

dimensional stress distribution. J. Prosthodont. 19, 425–431.

Albrektsson, T.A., 1988. Multicenter report on osseointegrated oral implants. J.

M

Prosthet. Dent. 60, 75-84.

d

Alhashim, A., Kamel, M., Brackett, W.W., 2012. Four-year follow-up of the

te

rehabilitation of a mandibular arch with a cementable zirconia-reinforced fixed dental

Ac ce p

prosthesis: A clinical report. J. Prosthet. Dent. 108, 138-142.

Bergendal, B., Palmqvist, S., 1999. Laser-welded titanium framework for implantsupported fixed prostheses: A 5-year report. Int. J. Oral Maxillofac. Implants. 14, 6971.

Bevilacqua, M., Tealdo, T., Menini, M., Pera, F., Mossolov, A., Drago, C. Pera, P., 2011. The influence of cantilever length and implant inclination on stress distribution in maxillary implant supported fixed dentures. J. Prosthetic Dent. 105, 5–13.

18

Page 18 of 41

Bickford, J.J.r., 1981. An introduction to the design and behavior of bolted joints. New York, Marcel Decker.

ip t

Byrne, D., Jacobs, S., O'Connell, B., Houston, F., Claffey, N., 2006. Preloads generated with repeated tightening in three types of screws used in dental implant

cr

assemblies. J. Prosthodont. 15, 164-171.

us

Cantwell, A., Hobkirk, J.A., 2004. Preload loss in gold prosthesisretaining screws as

an

a function of time. Int. J. Oral Maxillofac. Implants. 19, 124-132.

M

Carr, A.B., Gerard, D.A., Larsen, P.E., 1996. The response of bone in primates around unloaded dental implants supporting prostheses with different levels of fit. J.

te

d

Prosthetic Dent. 76, 500–509.

Choi, J.H., Lim, Y.J., Kim, C.W., Kim, M.J., 2009. The effect of different screw-

Ac ce p

tightening techniques on the stress generated on an internal-connection implant superstructure. Int. J. Oral Maxillofac. Implants. 24, 1045-1053.

Coelho, P.G., Sudack, P., Suzuki, M., Kurtz, K.S., Romanos, G.E. Silva, N.R.F.A., 2008. In vitro evaluation of the implant abutment connection sealing capability of different implant systems. J. Oral Rehabil. 35, 917-924.

Craig, R.G., Hanks, C.T., 1990. Cytotoxicity of experimental casting alloys evaluated by cell culture tests. J. Dent. Res. 69, 1539-1542.

19

Page 19 of 41

Curtis, D.A., Sharma, A., Finzen, F.C., Kao, R.T., 2000. Occlusal considerations for implant restorations in the partially edentulous patient. J. Calif. Dent. Assoc. 28, 771-

ip t

779.

Greco, G.D., Jansen, W.C., Landre Junior, J., Seraidarian, P.I., 2009. Biomechanical

cr

analysis of the stresses generated by different disocclusion patterns in an implant-

us

supported mandibular completedenture. J. Appl. Oral Science. 1, 515–520.

an

Greven, B., Luepke, M., von Dorche, S.H., 2007. Telescoping implant prostheses with intraoral luted galvano mesostructures to improve passive fit. J. Prosthet. Dent.

M

98, 239-244.

d

Griffith, H.T., 1987. Suggested tightening torques for structural bolts. Fastener

te

Technolog/February 1987. In: Torque Tensioning: A Ten Part Complication. Stow,

Ac ce p

OH, Fastener Technology.

Haack, J.E., Sakaguchi, R.L., Sun, T., Coffey, J.P., 1995. Elongation and preload stress in dental implant abutment screws. Int. J. Oral Maxillofac. Implants. 10, 529536.

Hagiwara, M., Ohashi, N., 1994. A new tightening technique for threaded fasteners. J. Offshore. Mech. Arct. 116, 64-69.

20

Page 20 of 41

Jacques, L.B., Moura, M.S., Suedam, V., Souza, E.A., Rubo, J.H., 2009. Effect of cantilever length and framework alloy on the stress distribution of mandibular-

ip t

cantilevered implant-supported prostheses. Clin. Oral Implants Res. 20, 737–741.

Jemt, T., Petersson, A., 1999. Precision of CNC-milled titanium frameworks for

cr

implant treatment in the edentulous jaw. Int. J. Prosthodont. 12, 209-215.

us

Jorneus, L., Jemt, T., Carlsson, L., 1992. Loads and designs of screw joints for

an

single crowns supported by osseointegrated implants. Int. J. Oral Maxillofac.

M

Implants. 7, 353-359.

Kansu, G., Aydin, A.K., 1996. Evaluation of the biocompatibility of various dental

te

d

alloys: Part I–Toxic potentials. Eur. J. Prosthodont. Restor. Dent. 4, 129-136.

Karl, M., Winter, W., Dickinson, A.J., Wichmann, M.G., Heckmann, S.M., 2007.

Ac ce p

Different bone loading patterns due to fixation of three-unit and five-unit implant prostheses. Aust. Dent. J. 52, 47–54.

Kim, Y., Oh, T.J., Misch, C.E., Wang, H.L., 2005. Occlusal considerations in implant therapy: Clinical guidelines with biomechanical rationale. Clin. Oral Implants Res. 16, 26-35.

Maló, P., de Araújo Nobre, M., Borges, J., Almeida, R., 2012. Retrievable metal ceramic implant-supported fixed prostheses with milled titanium frameworks and all-

21

Page 21 of 41

ceramic crowns: retrospective clinical study with up to 10 years of follow-up. J. Prosthodont. 21, 256-264.

ip t

Markarian, R.M., Ueda, C., Sendyk, C.L., Lagana, D.C. Souza, R.M., 2007. Stress distribution after installation of fixed frameworks with marginal gaps over angled and

cr

parallel implants: A photoelastic analysis. J. Prosthodont. 16, 117–122.

us

Martin, W.C., Woody, R.D., Miller, B.H., Miller, A.W., 2001. Implant abutment screw

an

rotations and preloads for four different screw materials and surfaces. J. Prosthet.

M

Dent. 86, 24-32.

McGlumpy, E.A., Mendel, D.A., Holloway, J.A., 1998. Implant screw mechanics.

te

d

Dent. Clin. North Am. 42, 71-89

Ac ce p

Mericske-Stern, R., 2008. Prosthetic considerations. Aust. Dent. J. 53, S49-S59.

Meleo, D., Baggi, L., Di Girolamo, M., Di Carlo, F., Pecci, R., Bedini, R., 2012. Fixture-abutment connection surface and micro-gap measurements by 3D microtomographic technique analysis. Ann. Ist. Super Sanita. 48, 53-58.

Örtorp, A., Jemt, T., 2009. Early laser-welded titanium frameworks supported by implants in the edentulous mandible: A 15-year comparative follow-up study. Clin. Implant Dent. Relat. Res. 11, 311-322.

22

Page 22 of 41

Ogawa, T., Dhaliwal, S., Naert, I., Mine, A., Kronstrom, M., Sasaki, K., Duyck, J., 2010. Impact of implant number, distribution and prosthesis material on loading on

ip t

implants supporting fixed prostheses. J. Oral Reh. 37, 525–531.

Oh, T.J., Yoon, J., Misch, C.E., Wang, H.L., 2002. The causes of early implant bone

cr

loss: Myth or science? J. Periodontol. 73, 322-333.

us

Park, J.K., Choi, J.U., Jeon, Y.C., Choi, K.S., Jeong, C.M., 2010. Effects of abutment

an

screw coating on implant preload. J. Prosthodont. 19, 458-464.

M

Patterson, E.A., Johns, R.B., 1992. Theoretical analysis of the fatigue life of fixture

d

screws in osseointegrated dental implants. Int. J. Oral Maxillofac. Implants. 7, 26-33.

te

Puri, S., 2005. Techniques used to fabricate all-ceramic restorations in the dental

Ac ce p

practice. Compend. Contin. Educ. Dent. 26, 519-525.

Quirynen, M., De Soete, M., van Steenberghe, D., 2002. Infectious risks for oral implants: A review of the literature. Clin. Oral Implants Res. 13, 1-19.

Rahimi, A., Keilig, L., Bendels, G., Klein, R., Buzug, T.M., Abdelgader, I., Abboud, M., Bourauel, C., 2005. 3D reconstruction of dental specimens from 2D histological images and µCT-scans. Comp. Meth. Biomech. Biomed. Eng. 8, 167-176.

23

Page 23 of 41

Rojas-Vizcaya, F., 2011. Full zirconia fixed detachable implant-retained restorations manufactured from monolithic zirconia: Clinical report after two years in service. J.

ip t

Prosthodont. 20, 570-576.

Rubo, J.H., Souza, E.A., 2008. Finite element analysis of stress in bone adjacent to

cr

dental implants. J. Oral Implantol. 34, 248–255.

Sadowsky, S.J., Caputo, A.A., 2000. Effect of anchorage systems and extension

us

base contact on load transfer with mandibular implant-retained overdentures. J.

an

Prosthet. Dent. 84, 327–334.

M

Sagat, G., Yalcin, S., Gultekin, B.A., Mijiritsky, E., 2010. Influence of arch shape and implant position on stress distribution around implants supporting fixed full- arch

te

d

prosthesis in edentulous maxilla. Implant Dent. 19, 498–508.

Sakaguchi, R.L., Sun, T., Haack, J.E., 1994. External strain distribution on implant

Ac ce p

prosthetic components. J. Dent. Res. 73, 232.

Segal, B.S., 2001. Prospective assessment of 546 all-ceramic anterior and posterior crowns in general practice. J. Prosthet. Dent. 85, 544-550.

Simsek, B., Erkmen, E., Yilmaz, D., Eser, A., 2006. Effects of different inter-implant distances on the stress distribution around endosseous implants in posterior mandible: A 3D finite element analysis. Med. Eng. Phys. 28, 199–213.

24

Page 24 of 41

Suedam, V., Souza, E.A., Moura, M.S., Jacques, L.B., Rubo, J.H., 2009. Effect of abutment’s height and framework alloy on the load distribution of mandibular

ip t

cantilevered implant-supported prosthesis. Clin. Oral Implants Res. 20, 196–200.

Tan, K.B., Nicholls, J.I., 2002. The effect of 3 torque delivery systems on gold screw

cr

preload at the gold cylinder-abutment screw joint. Int. J. Oral Maxillofac. Implants. 17, 175-183.

us

de Torres, E.M., Barbosa, G.A., Bernardes, S.R., de Mattos Mda, G., Ribeiro, R.F.,

an

2011. Correlation between vertical misfits and stresses transmitted to implants from

M

metal frameworks. J. Biomech. 44, 1735-1739.

Winter, W., Mohrle, S., Holst, S., Karl, M., 2010. Bone loading caused by different

d

types of misfits of implant-supported fixed dental prostheses: A three-dimensional

Ac ce p

Implants. 25, 947–952.

te

finite element analysis based on experimental results. Int. J. Oral Maxillofac.

25

Page 25 of 41

Tables Table 1: Components of the numerical model and their material properties. Material

Young’s modulus (MPa)

Poisson’s ratio

Cortical bone

20,000

0.30

Cancellous bone

1,000

0.30

ip t

Component

Implant

Titanium Grade V

110,000

0.30

Abutment

Titanium Grade V

110,000

Retention screw

Titanium Grade V

110,000

Cement (50 µm)

Zinc oxide eugenol

280

Coping layer 1 (0.30 mm)

Pure gold

79,000

0.44

Coping layer 2 (0.25 mm)

Hybrid composite

22,000

0.27

Framework 1

Gold alloy type III

92,000

0.32

Framework 2

Titanium Grade V

110,000

0.30

Framework 3

Zirconia

200,000

0.33

0.30

0.30

Ac ce p

te

d

M

an

us

cr

0.30

26

Page 26 of 41

ip t cr

us

Table 2: Obtained values for the maximal von Mises stress in MPa of the whole components of the FE models with the different applied loading conditions.

Coping layer 2 (hybrid composite)

Coping layer 1 (pure gold)

M an

Framework

Implant system

Cortical bone

Vertical 500 N, 2 points

Titanium alloy

511

153

266

460

69

Gold alloy

598

174

357

540

87

Zirconia

572

223

190

620

76

617

440

880

78

646

474

760

44

301

204

658

89

1,795

Gold alloy

1,757

Zirconia

1,203

ce pt

Titanium alloy

ed

Vertical 500 N,centre

30° load of 500 N, 2 points

1,441

376

403

620

92

Gold alloy

1,446

368

430

640

92

Zirconia

1,413

411

318

860

88

Titanium alloy Gold alloy Zirconia

Ac

Titanium alloy

2,418

442

309

841

113

2,376

460

329

850

89

2,488

512

233

970

115

450

112

950

100

30° load of 500 N, centre

950 (titanium grade 5) Yield strength 240 (gold alloy type III) 1000 (zirconia)

27 Page 27 of 41

ip t cr

applied loading conditions. Coping layer 2 (hybrid composite)

Coping layer 1 (pure gold)

M an

Framework

us

Table 3: Obtained values for the maximal equivalent total strain of the whole components of the FE models with the different

Implant system

Cancellous bone

Vertical 500 N, 2 points

Titanium alloy

0.0046

0.0018

0.0033

0.0023

3,409 µε

Gold alloy

0.0063

0.0047

0.0045

0.0030

4,000 µε

Zirconia

0.0031

0.0034

0.0025

0.0050

3,000 µε

0.0132

0.0066

0.0070

5,500 µε

0.0143

0.0071

0.0050

6,000 µε

0.0059

0.0031

0.0048

3,429 µε

0.0124

0.0090

0.0050

8,500 µε

0.0125

0.0091

0.0100

7,200 µε

0.0123

0.0089

0.0070

7,500 µε

0.0167

Gold alloy

0.0197

Zirconia

0.0063

0.0127

Gold alloy

0.0154

Zirconia

0.0098

Titanium alloy Gold alloy Zirconia

Ac

Titanium alloy

ce pt

Titanium alloy

ed

Vertical 500 N, centre

30° load of 500 N, 2 points

30° load of 500 N, centre

0.0232

0.0089

0.0061

0.0061

7,092 µε

0.0278

0.0093

0.0062

0.0070

6,000 µε

0.0138

0.0086

0.0063

0.0075

8,000 µε

28 Page 28 of 41

Legends for Figures Figure 1: a) Longitudinal cross-section of the whole model. b) Magnified crosssection of the framework at the connection part with the abutment with the different

ip t

coping layers.

cr

Figure 2: Loading conditions of the framework with a total force of 500 N:

Vertical loading condition: 1a) Loading of the framework at the implant-supporting

us

positions with a vertical force of 250 N at each position. 1b) Loading the centre of the

an

framework with 500 N applied vertically.

Angular Loading condition at 30° to the framework’s long axis in bucco-lingual

M

direction: a) Loading of the framework with a force of 250 N 30° to its long axis at the

d

implant-supporting positions. b) Loading the centre of the framework with 500 N.

te

Figure 3: a) Buccal view of stress distributions within the prosthesis and implant system under 500 N of vertical load (left) and 30° in bucco-lingual direction (right). b)

Ac ce p

Obtained maximum values of equivalent stress of the different loading conditions for the three investigated framework materials. The dashed lines represent the yield strength of the framework materials, namely, 1,000 MPa for zirconia, 240 MPa for gold alloy type III, and 950 MPa for titanium grade 5.

Figure 4: a) Buccal view of stress distributions within the prosthesis and the implant system under 500 N of vertical load (left) and 30° in bucco-lingual direction (right). b) Obtained maximum values of equivalent strain for the three investigated framework materials under different loading conditions.

29

Page 29 of 41

Figure 5: a) Buccal view of stress distributions within the cortical bone under 500 N of vertical load (left) and 30° in bucco-lingual direction (right). b) Obtained maximum values of equivalent stress for the cortical bone under different loading conditions.

ip t

The stress values are presented after the subtraction of those caused by the

cr

preloading of the retention screw.

Figure 6: a) Lingual view of strain distributions within the cancellous bone under 500

us

N vertical loads (left) and 30° in bucco-lingual direction (right). b) Obtained maximum

an

values of equivalent total strain for the cancellous bone under different loading

Ac ce p

te

d

M

conditions.

30

Page 30 of 41

Ac

ce

pt

ed

M

an

us

cr

i

Figure_1a

Page 31 of 41

Ac

ce

pt

ed

M

an

us

cr

i

Figure_1b

Page 32 of 41

Ac ce p

te

d

M

an

us

cr

ip t

Figure_2

Page 33 of 41

Ac ce p

te

d

M

an

us

cr

ip t

Figure_3a

Page 34 of 41

Ac

ce

pt

ed

M

an

us

cr

i

Figure_3b

Page 35 of 41

Ac ce p

te

d

M

an

us

cr

ip t

Figure_4a

Page 36 of 41

Ac

ce

pt

ed

M

an

us

cr

i

Figure_4b

Page 37 of 41

Ac ce p

te

d

M

an

us

cr

ip t

Figure_5a

Page 38 of 41

Ac

ce

pt

ed

M

an

us

cr

i

Figure_5b

Page 39 of 41

Ac ce p

te

d

M

an

us

cr

ip t

Figure_6a

Page 40 of 41

Ac

ce

pt

ed

M

an

us

cr

i

Figure_6b

Page 41 of 41