- Email: [email protected]
Evaluation of the accuracy and stress distribution of 3-unit implant supported prostheses obtained by different manufacturing methods
Materials Science & Engineering C 102 (2019) 66–74
Contents lists available at ScienceDirect
Materials Science & Engineering C journal homepage: www.elsevier.com/locate/msec
Evaluation of the accuracy and stress distribution of 3-unit implant supported prostheses obtained by different manufacturing methods
T
Caroline C. Melloa, Joel F. Santiago Juniorb, Cleidiel A.A. Lemosa, Graziella A. Galhanoc, ⁎ Edoardo Evangelistid, Roberto Scottid, Fellippo R. Verria, Eduardo P. Pellizzera, a
Department of Dental Materials and Prosthodontics, São Paulo State University (UNESP), School of Dentistry, Araçatuba, São Paulo, Brazil Department of Health Sciences, University of Sacred Heart - USC, São Paulo, Brazil c Department of Dental Materials and Prosthodontics, University of Western São Paulo, Presidente Prudente, São Paulo, Brazil d Department of Biomedical Sciences and Neuromotor, Alma Mater Studiorum University of Bologna, Italy b
A R T I C LE I N FO
A B S T R A C T
Keywords: Dental implant Prosthodontics CAD-CAM Marginal misfit Finite element analysis
The purpose of this in vitro study was to measure the vertical, positive-horizontal, and negative-horizontal misfit (VM, PHM, and NHM, respectively) of the zirconia three-element prosthetic framework, fabricated using different methods, and compare them with conventional fabrication methods (lost-wax casting). Furthermore, this study aimed to evaluate the influence of the misfit values on the biomechanical behavior of the 3-unit fixed prosthetic frameworks using three-dimensional finite element analysis (3D-FEA). Forty frameworks (n = 10) were fabricated as follow: G1, Cerec Bluecam; G2, iTero; G3, 3Series; and G4, conventional method. The samples were randomized to measure marginal misfit using a high-precision three-dimensional (3D)-optical microscope. The results were submitted to analysis of variance (ANOVA), with the significance level set at 5%. The mean VM values of each group were used in creating the models by 3D-FEA with the misfit found in optical microscopy. The programs used were the InVesalius, Rhinoceros, SolidWorks, FEMAP and NEiNastran. The von Mises map was plotted for each model. The G4 showed the lowest mean VM value (16.73 μm), followed by G3 (20.71 μm), G2 (21.01 μm), and G1 (41.77 μm) (p < 0.001). G2 was more accurate than G1 (p < 0.05) and similar to G3 (p = 0.319). For PHM, G4 was the most accurate and did not present overextended values. With regard to NHM, the computer-aided design and computer-aided manufacturing (CAD-CAM) systems were more accurate (−61.91 μm) than G4 (−95.36 μm) (p = 0.014). In biomechanical analysis, stress concentration caused by oblique loading is greater than caused by axial loading. In axial loading, G4 was the most favorable while G1 was the least favorable, biomechanically, in oblique loading, similar stress patterns were observed in all the models. The prosthetic screw was the most overloaded structure, but the material did not influence the stress distribution. The misfit prostheses showed a greater degree of stress than the controls (without misfit). The manufacturing method influenced the marginal misfit of the frameworks, with the conventional method being the most accurate and the Cerec Bluecam System (closed system) the least accurate. Biomechanically, fitting prostheses were more favorable than misfit prostheses.
1. Introduction In the last decades, the classical literature in the area of implantology has shown the importance of fabricating a well-adapted implant-supported prostheses and the biological and mechanical effects that can originate from a prosthetic misfit [1,2]. Thus, computer-aided design and computer-aided manufacturing (CAD-CAM) systems were introduced in the process of dental production during the 1970s, and specifically, to implantology in the mid-1990s, transforming the
manufacturing process of the implant abutments [3] and making it possible to automate the process of manufacturing prosthetic restorations. The insertion into the dental market of highly resistant and aesthetic materials that were exclusively developed using CAD-CAM technology [4], mainly allowed for greater accuracy in the preparation of restorations and prosthetic components [5–8]. These systems are divided into the extraoral (indirect) and the intraoral (direct) techniques [9]. Recent studies have verified a similar or better precision of the digital scans as that obtained by conventional
⁎ Corresponding author at: Department of Dental Materials and Prosthodontics, UNESP - São Paulo State University, José Bonifácio st., 1193, Vila Mendonça, Araçatuba, São Paulo 16050-050, Brazil. E-mail addresses: [email protected], [email protected] (E.P. Pellizzer).
https://doi.org/10.1016/j.msec.2019.03.059 Received 8 September 2018; Received in revised form 18 March 2019; Accepted 18 March 2019 Available online 19 March 2019 0928-4931/ © 2019 Published by Elsevier B.V.
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
suitable digital data from G1, a contrast powder (CEREC Optispray, Sirona) was applied under the resin matrix surface to be scanned. Ten other scans of the previous plaster models were made using the extraoral system with the aid of the same transfers used in the anterior step (Fig. 1). The data from Groups 2 and 3 were obtained in standard tessellation language (.STL) format and transferred to the Implant-Abutment Module software program (DWOS – Dental Wings) to design the G1, G2, and G3 implant-supported frameworks. The frameworks had standard thicknesses of 0.6 mm for the crown and 4 mm for the connectors, comprising multiple implants and without an anti-rotational system. The obtained images were displayed at different angles to observe any flaws and then transferred to milling on the Industrial Complex (Conexão Sistemas de Prótese) for the open groups (G2 and G3). For Group I, which is a closed system, the digital data were obtained as .inlab file formats and were milled in a machine from the same manufacturer (InLab MC XL, Sirona, Bensheim, Hessen, Germany) and using the same material used in the other groups (pre-sintered zirconia - VITA Zahnfabrik, Bad Säckingen, Germany). All groups in the CAD-CAM systems were milled in pre-sintered zirconia (VITA Zahnfabrik) held in a furnace (inFire HTC Speed, Sirona) at 1500 °C for 2 h and 20 min after the set identification of the ceramic material. For the manufacturing of the frameworks by the conventional method of lost-wax casting (G4, n = 10), 20 universal cast long abutments (UCLA) (Conexão Sistemas de Prótese Ltda.) with a pre-machined cylinder overcasting with cobalt-chromium (CoeCr) alloy were processed by the same technician under the same conditions. All the finished frameworks were taken for analysis of the vertical and horizontal marginal misfit. The 40 frameworks were screwed in the implant with a calibrated torque using a torquemeter (Conexão Sistemas de Prótese Ltda) under the resin matrix. The VM, PHM and NHM measurements were performed in a 3D optical microscope (Fig. 2) with a digital table, 350× magnification, and a precision of 1 μm. The measures were calculated using the QSPAK computer program (Mitutoyo). Nine equidistant and predetermined points were demarcated 1 mm below the implant platform in order to standardize the sequence and position the measurements. A total of 1440 measurements were performed and calculated as follow: 2 dental elements × (9 vertical points × 40 samples) + 2 dental elements × (9 horizontal points × 40 samples). For the analysis of the results, initially, a first examiner was responsible for the numbering of the 40 frameworks. Then, a second examiner performed the randomization of 40 numbers using Research Randomizer (Version 4.0). A single operator was responsible for all the measurements of marginal misfit and the data were obtained in micrometers. Twenty percent of the samples (n = 8) were re-read for an intra-examiner calculation to verify the calibration of the examiner and concordance between the analyses. With respect to statistical analyses, the experimental variables or between-subject factors were as follow: manufacturing technique (CADCAM systems x conventional technique), scanning system (intraoral system x extraoral system) and the use of the data acquisition (contrast powder x without contrast powder). The values of the measurements were submitted for statistical analysis using the SigmaPlot software program (SigmaPlot 13, Systat Software). The level of significance was 5%.
impression methods [10,11]. However, although different CAD-CAM systems are available for use, little is known about the most accurate scanning and production technologies used in the manufacturing of implant-supported components and restorations. These systems are being widely evaluated using a variety of methodologies, and studies have demonstrated a small value of marginal misfit with the use of the CAD-CAM technology [12,13]. However, studies on the correlation or real consequence that a prosthetic misfit can cause on the components involved in the implant-prosthesis components are scarce. Different methodologies are used to verify the biomechanical effects of prosthetic misfit [14–16]. The three-dimensional finite element analysis (3D-FEA) has been shown to be an effective tool for analyzing the biomechanical behaviors of prostheses and surrounding structures and has been used to evaluate the effect of a misfit on the biomechanical behavior of implant-supported prostheses [17–19]. Research on the scientific basis of the treatment or rehabilitation with implant-supported prosthetic restorations and components made by the CAD-CAM systems and stress distribution generated on the prosthetic components are of great importance. Thus, the aim of this study was to measure the vertical misfit (VM), positive horizontal misfit (PHM), and negative horizontal misfit (NHM) of frameworks made by different CAD/CAM methods and to compare them with the frameworks manufactured by the conventional method (lost-wax casting technique). In addition, the misfit results obtaining from each group was simulated in 3D-FEA to verify the misfit influence on stress distribution in the prosthetic components. The null hypotheses formulated were as follows: (1) the misfit frameworks do not present difference in the values of marginal misfit, independent of the manufacturing method used; and (2) the misfit implant-supported prostheses have no influence in the stress distribution. 2. Materials and methods 2.1. Analysis of vertical and horizontal misfit Four groups were obtained for VM and HM analyses, according to Table 1. Two external hexagon implants (Conexão Sistemas de Prótese) with a prosthetic platform of 4.1 mm in diameter and 10.0 mm in height were embedded in acrylic resin (JET, Clássico) for making a standard matrix and simulating a three-element, implant-supported prosthesis in the mandibular posterior region (1st premolar, 2nd premolar [pontic], and 1st molar) (Fig. 1). Twenty conventional matrix resin impressions were made by double mixture using polyvinyl siloxane (Elite H-D Putty Soft Normal Setting Zhermack) with an open technique with squared transfer copings, giving rise to 20 plaster models with die stone type IV (Elite Rock Thixotropic – Zhermack) to fabricate G3 (extraoral scanner, 3Series, DWOS Dental Wings) and G4 (lost-wax casting technique) frameworks (Fig. 1). Three different digital impression systems were used to obtain the digital data of the implants (Table 2). Twenty scans were performed using the resin matrix with intraoral systems (n = 10) and two specific transfers (Scan Body Unico – Simbiosi) for use with the CAD-CAM systems (G1: Cerec Bluecam, Sirona; and G2: iTero, Cadent), according to the guidelines recommended by each manufacturer. To obtain the Table 1 Experimental groups. Groups (n = 10)
Manufacturing systems
G1
Cerec BlueCam/Sirona (intraoral closed system + contrast powder) iTero/Cadent (intraoral open system without contrast powder) 3Series/DWOS (extraoral scanning system) Conventional method of lost-wax casting (control)
G2 G3 G4
2.2. Three-dimensional finite element analysis (3D-FEA) This methodology was performed following protocols from previous studies [20–22]. The mean of marginal misfit data measured in the previous methodology were calculated according to each point measured throughout the perimeter of each framework in each group. These mean of misfit data were used to simulation of the 3D-FEA misfit model for each group (G1, G2, G3 and G4) to evaluate biomechanical behavior 67
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 1. Illustrative image of the matrix and the implants that were used and all steps for frameworks fabrication of each group. Nine demarcated points around the implant platform to standardize the measurement regions of the misfit.
of implants, screw fixation and frameworks. According to Table 3, six 3D models (one for each group, and two control without marginal misfit and four models with marginal misfit data mentioned above) were simulated. Each model represent a bone block section (cortical and trabecular), with two external hexagon implants (4.0 × 10 mm), supporting a three-unit fixed prostheses (1st premolar, 2nd premolar [pontic] and 1st molar) with metal-ceramic or metal-free screwed crowns. The bone block was obtained from a computed tomography image (in .dicom format) using Invesalius 3.0.0., (CTI Renato Archer) simulating a bone with a layer of cortical bone thickness of 1 mm surrounding a nucleus of low-density trabecular bone [23]. The implant design and prosthetic components (UCLA abutment) were obtained from original drawings. The crowns were obtained through a surface scanning of a dental mannequin using a 3D scanner (Scanner 3D MDX20, Roland DG). All structures were exported to the Rhinoceros 3D 4.0. program (NURBS Modeling for Windows), for simplification of solids, but maintained the same characteristics as the originals. The whole set (implant/abutment/crown) was incorporated into the previously modeled bone block for further export of the solids to the finite element program [Fig. 3A–G]. After the modeling phase, all geometries were imported into the Femap 11.2 (Siemens PLM Software Inc., Santa Ana, CA, USA) for preand post-processing. The FEMAP was used to create finite element models from meshes of tetrahedral parabolic solid elements [Fig. 3H,
J]. All mechanical properties of each simulated material were attributed to the meshes according to previously published studies [17,24–26] (Table 4). For this analysis, all the materials involved in the study were considered homogeneous, isotropic, and linear. Furthermore, symmetric welds were simulated for all contacts, except for the contact between implant-abutment connections, which was simulated by symmetric contact. Constraint conditions were fixed in all axes (x, y, and z) at the boneblock sections (anterior and posterior surfaces), while all other parts in the models were unrestricted. The applied force were a total of 400 N in the axial direction (points determined on the internal slope of each cusps, totaling 50 N for cusps) and 200 N in the oblique direction (45° in relation to the long axis of the implant, only for internal slope of the buccal cusps) [27–30] (Fig. 3I). After the final elaboration of the finite element models, load incorporation, and movement restriction, they were exported to NEi Nastran 11 program (Nei Software) to analyze the previously modeled mathematical problem. After resolving the mathematical calculations in the NEi Nastran 11 program (Nei Software, Westminster, California, USA), the models were reimported into the Femap to visualize the results using the von Mises stresses (vM) maps with Megapascal (MPa) units.
Table 2 Selected characteristics of the scanning systems used in the present study, as provided by the respective manufacturer. Token
System
Manufacturer
Measurement principle
Measurement type
Measurement prerequisites
Output format
CEREC
CEREC Ac Bluecam
Sirona Dental Systems
Active triangulation
Powdering
.inlab
ITE
iTero
Cadent
Parallel confocal imaging
–
.stl
3S
3Series
Dental Wings Open Software
Laser triangulation to three axis of freedom
Still images. Accuracy of 15 μm Still images. Accuracy of 15 μm Still images. Accuracy < 20 μm
–
.stl
68
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 2. (A) Matrix with the demarcated measurement points on each implant. (B) Device/matrix/framework set under the microscope's digital table for further analysis of misfits. (C) Illustrative image of the computer program connected to the microscope, which were used for the measurements. (D) Example of a vertical marginal misfit region. (E) Example of a horizontal marginal misfit region.
Table 3 Models specifications. Models
Description
C1 C2 G1 G2 G3 G4
CONTROL - Completed fit framework manufacturing with Y-TZP (Yttria Tetragonal Zirconia Polycrystal). CONTROL - Completed fit framework manufacturing with NieCr (nickel chromium alloy). Misfit framework manufacturing with Y-TZP based on the misfit values of the G1 group frameworks (Cerec Bluecam CAD-CAM system - Sirona). Misfit framework manufacturing with Y-TZP based on the misfit values of the G2 group frameworks (iTero CAD-CAM system - Cadent). Misfit framework manufacturing with Y-TZP based on the misfit values of the G3 group frameworks (3Series CAD-CAM system - DWOS). Misfit framework manufacturing with NieCr based on the misfit values of the G4 group frameworks (Conventional Method of lost-wax casting).
Fig. 3. Schematic view of simulated model (A: final model; B: crowns; C: fixation screw; D: frameworks; E: dental implants; F: cortical bone; G: trabecular bone), meshes (H: lateral view; J: occlusal view); loading applied (I). Table 4 Mechanical properties of simulated materials. Structure
Elastic modulus (E) (GPa)
Poisson coefficient (ν)
References
Low-density trabecular bone (type III bone) Cortical bone Titanium (abutment, implant) Ni-Cr alloy Zirconia Feldspathic porcelain
1,37 13,7 110,0 200,0 205,0 82,8
0,30 0,30 0,35 0,33 0,22 0,35
Sevimay et al. [38] Sertgoz. [39] Sertgoz. [39] Pierrisnard et al. [40] Coelho et al. [41] Papavasiliou et al. [42]
3. Results
statistically more accurate (p < 0.001) than the CAD-CAM systems (Fig. 4A). In a specific analysis, the G1 (Cerec Bluecam) group had the highest mean value of VM at 41.77 μm, followed by 21.01 μm for G2 (iTero), 20.71 μm for G3 (3Series), and 16.73 μm for G4 (conventional) (Fig. 4B). When comparing the G1 with the G2, it was observed that the mean G1 values were statistically higher than the G2 values (p < 0.05) (Fig. 4B). In the comparative analysis between G2 and G3, it was observed that there was no significant difference between the groups (p = 0.319), in contrast to what was observed in the comparative analysis between G3 and the G1 (p < 0.001) (Fig. 4B). When comparing the PHM values of the CAD-CAM systems versus the conventional method, it was observed that the conventional method
3.1. Marginal misfit analysis Based on the systematic error, the intra-examiner analysis (before and after values of 20% of the frameworks, n = 8), indicated that there was no significant difference in the before and after analysis (VM = paired t-test: 1.563; p = 0.120 e HM = paired t-test: 0.829; p = 0.408). The random error coefficient (Dahlberg) indicated error margins of 10 μm and 2 μm for the values of vertical and horizontal marginal misfit, respectively. For the VM, the comparison of the CAD-CAM systems versus the conventional method showed that the conventional method was
69
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Table 5 Mean values of vertical marginal misfit (μm), of each point in each group, obtained by optical microscopy and used in the generation of finite element models. Fig. 4. Graphs of comparative analysis (A: CAD-CAM systems and conventional method for VMD; B: Difference between the groups evaluated for VMD; C: Analysis for NHM of CAD).
Points
1 2 3 4 5 6 7 8 9
Group 1
Group 2
Group 3
Group 4
PM
M
PM
M
PM
M
PM
M
44 43 32 37 30 38 44 42 35
57 60 51 36 36 36 45 42 43
46 32 18 12 9 19 24 24 38
12 17 23 18 22 21 22 13 9
24 25 14 16 11 23 25 25 13
16 31 28 23 26 22 23 15 12
22 15 18 25 26 32 25 16 14
11 14 14 6 7 13 16 17 12
was more accurate than the CAD-CAM systems, and it did not present any misfit related to this subject (over-contour). In the comparative analysis between the different intraoral scanning systems G1 and G2, no significant difference was observed between the groups (G1: 128.750 μm; G2 = 126.46 μm) (p = 0.734); however, when compared (G2 = 126.46 μm) with (G3: 116.79 μm) significant difference was observed (p = 0.018). For the NHM, the CAD-CAM systems (general average: −61.91 μm) were statistically more accurate than the conventional method (G4: −95.36 μm) (p = 0.014) (Fig. 4C). Among the CAD-CAM systems evaluated, the most accurate system was G2, which did not present NHM values, followed by G3, which presented only two values of NHM.
3.2. Finite element analysis The values of marginal misfit (μm) for each measured point obtained by in vitro study were used to simulate the 3D models (Table 5). Under axial loading, it is observed that the frameworks material did not influence the general stress distribution pattern of the misfit models. In the individual analysis of the frameworks, seen through an internal view, it can be observed that the completed fit models presented a uniform stress distribution with low stress values (6 to 75 MPa) in the cervical region of each framework. However, the models with misfit values present a high stress in the region of misfit (16 to 200 MPa) with low stress concentration in the G1 (highest VM) closer to the stress distribution of the control groups (Fig. 5). In the sagittal view of the prosthetic components, it was observed that the control groups models (C1 and C2) presented similar stress distribution with values lower than 75 MPa, whereas they showed higher stress concentrations among the misfit models, especially in the screw fixation of first molar (approximately 175 MPa). The Cerec Bluecam (G1, closed intraoral system with contrast powder) system model demonstrated a higher stress concentration in the fixation screw, followed by the 3Series (G3, open extraoral and powder free), iTero (G2, open intraoral and without contrast powder), and conventional methods (G4) (Figs. 6 and 7). Under oblique loading, all the models showed higher stress concentrations in the oblique loading when compared with the axial loading. The control groups were slightly more favorable than the misfit models, which presented greater stress at the lingual face of the cervical area of frameworks (Fig. 7). The Cerec Bluecam scanning system (G1, intraoral, closed, and with contrast powder) showed a slightly higher stress intensity and concentration than the other misfit models (Fig. 7). In the analysis of prosthetic components in sagittal view, the stress distribution pattern of all the models was similar to each other, with higher intensities of stress (150 to 450 MPa) detected in the fixation screws (Figs. 7 and 8).
70
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 5. von Mises stress maps: internal view of the frameworks. Axial and oblique loading. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
Fig. 6. von Mises stress maps: sagittal cut (anteroposterior), lingual view of prosthetic structures. Axial load. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
4. Discussion
the lowest VM values when compared with the misfit of the frameworks made by the CAD-CAM systems. Although many studies have observed high levels of accuracy of the CAD-CAM systems in implantology [5–8], the superiority of the conventional method determined in this study may be related to the fact that pillars with metal bases were used in lost-wax casting, which has its base mechanically made using computer numerical control in a similar way that the dental implants are made. Thus, it optimizes the
The creation of implant-supported dentures through CAD-CAM systems has become a practical and highly predictable practice for clinicians. The association of the two methodologies carried out in this study demonstrated that the first null hypothesis formulated was rejected because it was observed through optical microscopic measurements that the prostheses made by the conventional method obtained 71
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 7. von Mises stress maps: vestibular view of individualized prosthetic screws. Axial and oblique loading. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
Fig. 8. von Mises stress maps: sagittal cut (anteroposterior), lingual view of prosthetic structures. Oblique loading. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
of frameworks made by an intraoral scanning system (G2 [iTero]) and an extraoral scanning system (G3 [3Series]) (p = 0.319), which were also very close to the values obtained with the use of the conventional method (G4). At this point the literature is still not very clear, as some authors [32] believe that intraoral scanning is capable of offering greater precision, whereas others point out that the accuracy of extraoral scanning systems may be directly related to the technology used in
accuracy of adaptation between the abutment and platform of the implant, minimizing the misfits between components. Moreover, as observed in the study by Kahramaoglu & Kulak-Ozkan [31], the different values of the thermal expansion coefficient of zirconia, metallic alloy of the prosthetic component (CreCo), and metallic infrastructure nickelchromium (NieCr) may have influenced the results of this study. In our study, similar values of VM were observed in the comparison 72
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
movements [49], in addition to all the clinical procedures, have been performed under ideal laboratory conditions. The results observed by FEA analysis permit the understanding of biomechanical behavior of different structures; however, the results should be interpreted carefully due to limitation of computation simulation, such as materials properties are assumed to be isotropic, homogeneous, linear and static occlusal loading [16,20]. So, these limitations must be kept in mind before of extrapolated to the clinic day. However, when combined with trial clinical studies, these biomechanical analyses could serve as a guideline for use of the clinicians in implant dentistry. Therefore, further randomized-controlled trials should be conducted to gain a better understanding of the influence of marginal misfit in long term follow-up o implant-supported prostheses.
the scanning camera, as well as the optical quality of the pulverized material, such as contrast powder when used, and the resolution of the surface scanning device [33,34]. When comparing the two intraoral scanning systems used in this study, we observed that the system requiring the application of a contrast powder had the highest mean VM values (p < 0.05). Although the application of the powder facilitates the surface scanning, it can introduce a certain imprecision of scanning, due to the variation of its thickness (5 μm to 15 μm) [35]. However, it should be noted that all VM values reported in this study are within the clinically acceptable limits (i.e., 100 to 150 μm), to avoid mechanical and biological complications [36–38]. With regards to the PHM, the conventional method was more accurate and did not present frameworks with overextension or over configuration, whereas the CAD-CAM systems had lower average values of NHM than the conventional method. The frameworks of pre-sintered zirconia, on average, are milled 20% higher than the planned final dimensions. Before starting the milling procedures, a bar code reading in each ceramic block is performed, in which the milling software identifies the contraction rate of each material after being submitted to the sintering process [39–41]. For this reason, the results determined in this study may be related to this dimensional alteration encountered in the zirconia frameworks during the sintering, or to the fact that manual adjustments were not made in any part of this study. The second null hypothesis formulated in this study was also rejected. Using 3D-FEA, we evaluated the biomechanical behavior of the implant-prosthesis set, and we observed that prosthetic misfit influenced the stress distribution of the components evaluated. However, in this study, the frameworks of manufacturing material and small variations of misfit between the groups did not change the distribution pattern and the concentration of stress, corroborating previous studies [14,42,43]. There were no significant differences between the models with oblique loading. However, with axial loading, the conventional method (G1) presented the best biomechanical situation, which presented the lowest marginal misfit values. Vertical misfits are mainly related to, among other factors, fractures of the prosthetic components or frameworks, abutment and/or screw loosening. Misfits are associated with increased difficulty in achieving satisfactory dental hygiene. Consequently, they may induce greater retention of food and cause a greater bacterial accumulation [44], altering the buccal microflora and increasing the possibilities of the development of inflammatory reactions and infections in the peri-implant region, and may even increase the rate of marginal bone loss [42,45]. Although research studies strive to determine the clinically acceptable values of marginal misfits, little is known about the borderline misfit that does not significantly impair bone tissue and osseointegration of the implant. Thus, understanding the difficulty involved in obtaining an absolute passive settlement of the implant-supported components and restorations, most authors recommend minimizing, as much as possible, the misfits to avoid the development of the previously-mentioned possible complications [45–47]. Another recommendation is to establish a minimum passive settling of the components, in which the formation of static charges and high intensity stresses inside the prosthesis or in the surrounding bone tissue will not be induced [45,48]. Finally, from this study, we can consider that although the conventional method presented the least values of marginal misfit, both the intraoral and extraoral scanning technique can be predicted in the manufacturing of implant-supported prostheses, presenting frameworks with the best possible adaptation and optimizing the longevity of rehabilitation treatments. However, the accelerated development and improvement of the CAD-CAM systems in dentistry must be accompanied by constant professional training [50]. The limitations to this study are inherent to the in vitro methodologies that were used. The preparation of a matrix differs from a natural intraoral scan situation involving soft tissue, saliva, and patient
5. Conclusions Based on the methodologies and within the limitations of this in vitro study, the following conclusions were drawn: 1. The manufacturing method used influenced the misfit of the frameworks, for both vertical and horizontal measurements. With the lowest means of misfit, the conventional method being the most favorable and the Cerec Bluecam System the least favorable; 2. The frameworks misfit influenced the stress distribution in the prosthetic components, while the completely fit prostheses were more biomechanically favorable than the misfit prostheses. Acknowledgments Supported by (Scholarship) CNPq: 165406/2015-1 and São Paulo Research Foundation (FAPESP) Grant: #2009/16164-7, #2011/071347 and #2011/19150-7. References [1] E.A. Patterson, R.B. Johns, Theoretical analysis of the fatigue life of fixture screws in osseointegrated dental implants, Int. J. Oral Maxillofac. Implants 7 (1) (1992 Spring) 26–33. [2] W.G. Assunção, E.A. Gomes, E.P. Rocha, J.A. Delben, Three-dimensional finite element analysis of vertical and angular misfit in implant-supported fixed prostheses, Int. J. Oral Maxillofac. Implants 26 (2011) 788–796. [3] G. Priest, Virtual-designed and computer-milled implant abutments, J. Oral Maxillofac. Surg. 63 (2005) 22–32. [4] A. Syrek, G. Reich, D. Ranftl, C. Klein, B. Cerny, J. Brodesser, Clinical evaluation of all-ceramic crowns fabricated from intraoral digital impressions based on the principle of active wavefront sampling, J. Dent. 38 (2010) 553–559. [5] P. Vigolo, F. Fonzi, Z. Majzoub, G. Cordioli, An in vitro evaluation of titanium, zirconia, and alumina procera abutments with hexagonal connection, Int. J. Oral Maxillofac. Implants 21 (2006) 575–580. [6] P. Coli, S. Karlsson, Precision of a CAD-CAM technique for the production of zirconium dioxide copings, Int. J. Prosthodont. 17 (2004) 577–580. [7] I. Naert, A. Van der Donck, L. Beckers, Precision of fit and clinical evaluation of allceramic full restorations followed between 0.5 and 5 years, J. Oral Rehabil. 32 (2005) 51–57. [8] D.G. de França, M.H. Morais, F.D. das Neves, A.F. Carreiro, G.A. Barbosa, Precision fit of screw-retained implant-supported fixed dental prostheses fabricated by CADCAM, copy-milling, and conventional methods, Int. J. Oral Maxillofac. Implants (2016), https://doi.org/10.11607/jomi.5023 (Epub ahead of print). [9] H. Rudolph, H. Salmen, M. Moldan, K. Kuhn, V. Sichwardt, B. Wöstmann, et al., Accuracy of intraoral and extraoral digital data acquisition for dental restorations, J. Appl. Oral Sci. 24 (2016) 85–94. [10] F.S. Andriessen, D.R. Rijkens, W.J. van der Meer, D.W. Wismeijer, Applicability and accuracy of an intraoral scanner for scanning multiple implants in edentulous mandibles: a pilot study, J. Prosthet. Dent. 111 (2014) 186–194. [11] W.J. van der Meer, F.S. Andriessen, D. Wismeijer, Y. Ren, Application of intra-oral dental scanners in the digital workflow of implantology, PLoS One 7 (2012) e43312. [12] M.F. Leifert, M.M. Leifert, S.S. Efstratiadis, T.J. Cangialosi, Comparasion of space analysis evaluations with digital models and plaster dental casts, Am. J. Orthod. Dentofac. Orthop. 136 (2009) 16.e1–16.e4. [13] H.A. Mously, M. Finkkelman, R. Zandparsa, H. Hirayama, Marginal and internal adaptation of ceramic crown restorations fabricated with CAD-CAM technology and the heat-press technique, J. Prosthet. Dent. 112 (2014) 249–256. [14] J. Abduo, K. Lyons, Effect of vertical misfit on strain within screw-retained implant titanium and zirconia frameworks, J. Prosthodont. Res. 56 (2012) 102–109.
73
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
[15] D.K. de Vasconcellos, A.N. Kojima, A.M. Mesquita, M.A. Bottino, M. Ozcan, A microstrain comparison of passively fitting screw-retained and cemented titanium frameworks, J. Prosthet. Dent. 112 (4) (2014) 834–838. [16] F.R. Verri, J.F. Santiago Jr., D.A. Almeida, V.E. de Souza Batista, C.A. Araujo Lemos, C.C. Mello, et al., Biomechanical three-dimensional finite element analysis of single implant-supported prostheses in the anterior maxilla, with different surgical techniques and implant types, Int. J. Oral Maxillofac. Implants 32 (2017) e191–e198. [17] J. Abduo, R.B. Judge, Implications of implant framework misfit: a systematic review of biomechanical sequelae, Int. J. Oral Maxillofac. Implants 29 (2014) 608–621. [18] C. Manzella, C. Bignardi, V. Burello, S. Carossa, G. Schierano, Method to improve passive fit of frameworks on implant-supported prostheses: an in vitro study, J. Prosthet. Dent. 116 (2016) 52–58. [19] M.B. dos Santos, A. Bacchi, R.L. Consani, L. Correr-Sobrinho, Influence of diamondlike carbon coating of screws on axial tightening force and stress distribution on overdenture bar frameworks with different fit levels and materials, Int. J. Oral Maxillofac. Implants 30 (5) (2015 Sep-Oct) 1019–1027 (30). [20] E.P. Pellizzer, C.A.A. Lemos, D.A.F. Almeida, V.E. de Souza Batista, J.F. Santiago Júnior, F.R. Verri, Biomechanical analysis of different implant-abutments interfaces in different bone types: an in silico analysis, Mater. Sci. Eng. C Mater. Biol. Appl. 90 (2018) 645–650 Sep 1. [21] L. Minatel, F.R. Verri, G.A.H. Kudo, D.A. de Faria Almeida, V.E. de Souza Batista, C.A.A. Lemos, E.P. Pellizzer, J.F. Santiago, Junior. Effect of different types of prosthetic platforms on stress-distribution in dental implant-supported prostheses, Mater. Sci. Eng. C Mater. Biol. Appl. 71 (2017) 35–42 Feb 1. [22] J.F. Santiago Junior, F.R. Verri, D.A. Almeida, V.E. de Souza Batista, C.A. Lemos, E.P. Pellizzer, Finite element analysis on influence of implant surface treatments, connection and bone types, Mater. Sci. Eng. C Mater. Biol. Appl. 63 (2016) 292–300 Jun. [23] U. Lekholm, G.A. Zarb, P.-I. Brånemark, G.A. Zarb, T. Albrektsson, Tissue-integrated prostheses, Tissue Integrated Prostheses, Quintessence, Chicago, 1985, pp. 199–209. [24] M. Sevimay, F. Turhan, M.A. Kiliçarslan, G. Eskitascioglu, Three-dimensional finite element analysis of the effect of different bone quality on stress distribution in an implant-supported crown, J. Prosthet. Dent. 93 (2005) 227–234. [25] A. Sertgoz, Finite element analysis study of the effect of superstructure material on stress distribution in an implant supported fixed prosthesis, Int. J. Prosthodont. 10 (1997) 19–27. [26] P.G. Coelho, E.A. Bonfante, N.R. Silva, E.D. Rekow, V.P. Thompson, Laboratory simulation of Y-TZP all-ceramic crown clinical failures, J. Dent. Res. 88 (2009) 382–386. [27] C.A.A. Lemos, F.R. Verri, J.F. Santiago Junior, V.E. de Souza Batista, D.T. Kemmoku, P.Y. Noritomi, E.P. Pellizzer, Splinted and nonsplinted crowns with different implant lengths in the posterior maxilla by three-dimensional finite element analysis, J. Healthc. Eng. 2018 (2018 Sep 3) 3163096, https://doi.org/10. 1155/2018/3163096. [28] V.E. de Souza Batista, F.R. Verri, D.A.F. Almeida, J.F. Santiago Junior, C.A.A. Lemos, E.P. Pellizzer, Evaluation of the effect of an offset implant configuration in the posterior maxilla with external hexagon implant platform: a 3-dimensional finite element analysis, J. Prosthet. Dent. 118 (3) (2017) 363–371 Sep. [29] C.A.A. Lemos, F.R. Verri, Santiago J.F. Júnior, D.A.F. Almeida, V.E.S. Batista, P.Y. Noritomi, E.P. Pellizzer, Retention system and splinting on Morse taper implants in the posterior maxilla by 3D finite element analysis, Braz. Dent. J. 29 (1) (2018) 30–35 Jan-Feb. [30] V.E. de Souza Batista, F.R. Verri, D.A. Almeida, J.F. Santiago Junior, C.A. Lemos, E.P. Pellizzer, Finite element analysis of implant-supported prosthesis with pontic and cantilever in the posterior maxilla, Comput. Meth. Biomech. Biomed. Eng. 20 (6) (2017) 663–670 May. [31] E. Kahramanoğlu, Y. Kulak-Ozkan, Marginal and internal adaptation of different superstructure and abutment materials using two different implant systems for five-
[32]
[33] [34] [35]
[36]
[37] [38]
[39] [40]
[41]
[42]
[43]
[44]
[45]
[46] [47]
[48]
[49]
[50]
74
unit implant-supported fixed partial dentures: an in vitro study, Int. J. Oral Maxillofac. Implants 28 (2013) 1207–1216, https://doi.org/10.11607/jomi.2744. T.V. Flügge, S. Schlager, K. Nelson, S. Nahles, M.C. Metzger, Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a model scanner, Am. J. Orthod. Dentofac. Orthop. 144 (2013) 471–478, https://doi. org/10.1016/j.ajodo.2013.04.017. M. Persson, M. Andersson, B. Bergman, The accuracy of a high-precision digitizer for CAD-CAM of crowns, J. Prosthet. Dent. 74 (1995) 223–229. H.H. Zaghloul, J.F. Younis, Marginal fit of implant-supported all-ceramic zirconia frameworks, J. Oral Implantol. 39 (2013) 417–424. J. Katsoulis, R. Mericske-Stern, D.M. Yates, N. Izutani, N. Enkling, M.B. Blatz, In vitro precision of fit of computer-aided design and computer-aided manufacturing titanium and zirconium dioxide bars, Dent. Mater. 29 (2013) 945–953. T. Jemt, Failures and complications in 391 consecutively inserted fixed prostheses supported by Brånemark implants in edentulous jaws: a study of treatment from the time of prosthesis placement to the first annual checkup, Int. J. Oral Maxillofac. Implants 6 (1991) 270–276. T. Jemt, J.E. Rubenstein, L. Carlsson, B.R. Lang, Measuring fit at the implant prosthodontic interface, J. Prosthet. Dent. 75 (1996) 314–325. E. Bayramoğlu, Y.K. Özkan, C. Yildiz, Comparison of marginal and internal fit of press-on-metal and conventional ceramic systems for three- and four-unit implantsupported partial fixed dental prostheses: an in vitro study, J. Prosthet. Dent. 114 (2015) 52–58. F. Komine, M.B. Blatz, H. Matsumura, Current status of zirconia-based fixed restorations, J. Oral Sci. 52 (2010) 531–539. F. Filser, P. Kocher, F. Weibel, H. Lüthy, P. Schärer, L.J. Gauckler, Reliability and strength of all-ceramic dental restorations made by direct ceramic machining (DCM), Int. J. Comput. Dent. 4 (2001) 89–106. D. Suttor, K. Bunke, S. Hoescheler, H. Hauptmann, G. Hertlein, LAVA - the system for all-ceramic ZrO2 crown and bridge frameworks, Int. J. Comput. Dent. 4 (2001) 195–206. C. Kunavisarut, L.A. Lang, B.R. Stoner, D.A. Felton, Finite element analysis on dental implant-supported prostheses without passive fit, J. Prosthodont. 11 (2002) 30–40. R.T. Abreu, A.O. Spazzin, P.Y. Noritomi, R.L. Consani, M.F. Mesquita, Influence of material of overdenture-retaining bar with vertical misfit on three-dimensional stress distribution, J. Prosthodont. 19 (2010) 425–431. F.D. Neves, C.J. Prado, M.S. Prudente, T.A. Carneiro, K. Zancopé, L.R. Davi, et al., Micro-computed tomography evaluation of marginal fit of lithium disilicate crowns fabricated by using chairside CAD-CAM systems or the heat-pressing technique, J. Prosthet. Dent. 112 (2014) 1134–1140. M.M. Buzayan, N.B. Yunus, Passive fit in screw retained multi-unit implant prosthesis understanding and achieving: a review of the literature, J. Indian Prosthodont. Soc. 14 (2014) 16–23. S. Sahin, M.C. Cehreli, The significance of passive framework fit in implant prosthodontics: current status, Implant. Dent. 10 (2001) 85–92. S.A. de Sousa, M.A. de Arruda Nobilo, G.E. Henriques, M.F. Mesquita, Passive fit of frameworks in titanium and palladium-silver alloy submitted the laser welding, J. Oral Rehabil. 35 (2008) 123–127. R. Sorrentino, E.F. Gherlone, G. Calesini, F. Zarone, Effect of implant angulation, connection length, and impression material on the dimensional accuracy of implant impressions: an in vitro comparative study, Clin. Implant. Dent. Relat. Res. 12 (Suppl. 1) (2010) e63–e76. S.Y. Kim, M.J. Kim, J.S. Han, I.S. Yeo, Y.J. Lim, H.B. Kwon, Accuracy of dies captured by an intraoral digital impression system using parallel confocal imaging, Int. J. Prosthodont. 26 (2013) 161–163. G.Á. Galhano, E.P. Pellizzer, J.V. Mazaro, Optical impression systems for CAD-CAM restorations, J. Craniofac. Surg. 23 (2012) e575–e579.
Contents lists available at ScienceDirect
Materials Science & Engineering C journal homepage: www.elsevier.com/locate/msec
Evaluation of the accuracy and stress distribution of 3-unit implant supported prostheses obtained by different manufacturing methods
T
Caroline C. Melloa, Joel F. Santiago Juniorb, Cleidiel A.A. Lemosa, Graziella A. Galhanoc, ⁎ Edoardo Evangelistid, Roberto Scottid, Fellippo R. Verria, Eduardo P. Pellizzera, a
Department of Dental Materials and Prosthodontics, São Paulo State University (UNESP), School of Dentistry, Araçatuba, São Paulo, Brazil Department of Health Sciences, University of Sacred Heart - USC, São Paulo, Brazil c Department of Dental Materials and Prosthodontics, University of Western São Paulo, Presidente Prudente, São Paulo, Brazil d Department of Biomedical Sciences and Neuromotor, Alma Mater Studiorum University of Bologna, Italy b
A R T I C LE I N FO
A B S T R A C T
Keywords: Dental implant Prosthodontics CAD-CAM Marginal misfit Finite element analysis
The purpose of this in vitro study was to measure the vertical, positive-horizontal, and negative-horizontal misfit (VM, PHM, and NHM, respectively) of the zirconia three-element prosthetic framework, fabricated using different methods, and compare them with conventional fabrication methods (lost-wax casting). Furthermore, this study aimed to evaluate the influence of the misfit values on the biomechanical behavior of the 3-unit fixed prosthetic frameworks using three-dimensional finite element analysis (3D-FEA). Forty frameworks (n = 10) were fabricated as follow: G1, Cerec Bluecam; G2, iTero; G3, 3Series; and G4, conventional method. The samples were randomized to measure marginal misfit using a high-precision three-dimensional (3D)-optical microscope. The results were submitted to analysis of variance (ANOVA), with the significance level set at 5%. The mean VM values of each group were used in creating the models by 3D-FEA with the misfit found in optical microscopy. The programs used were the InVesalius, Rhinoceros, SolidWorks, FEMAP and NEiNastran. The von Mises map was plotted for each model. The G4 showed the lowest mean VM value (16.73 μm), followed by G3 (20.71 μm), G2 (21.01 μm), and G1 (41.77 μm) (p < 0.001). G2 was more accurate than G1 (p < 0.05) and similar to G3 (p = 0.319). For PHM, G4 was the most accurate and did not present overextended values. With regard to NHM, the computer-aided design and computer-aided manufacturing (CAD-CAM) systems were more accurate (−61.91 μm) than G4 (−95.36 μm) (p = 0.014). In biomechanical analysis, stress concentration caused by oblique loading is greater than caused by axial loading. In axial loading, G4 was the most favorable while G1 was the least favorable, biomechanically, in oblique loading, similar stress patterns were observed in all the models. The prosthetic screw was the most overloaded structure, but the material did not influence the stress distribution. The misfit prostheses showed a greater degree of stress than the controls (without misfit). The manufacturing method influenced the marginal misfit of the frameworks, with the conventional method being the most accurate and the Cerec Bluecam System (closed system) the least accurate. Biomechanically, fitting prostheses were more favorable than misfit prostheses.
1. Introduction In the last decades, the classical literature in the area of implantology has shown the importance of fabricating a well-adapted implant-supported prostheses and the biological and mechanical effects that can originate from a prosthetic misfit [1,2]. Thus, computer-aided design and computer-aided manufacturing (CAD-CAM) systems were introduced in the process of dental production during the 1970s, and specifically, to implantology in the mid-1990s, transforming the
manufacturing process of the implant abutments [3] and making it possible to automate the process of manufacturing prosthetic restorations. The insertion into the dental market of highly resistant and aesthetic materials that were exclusively developed using CAD-CAM technology [4], mainly allowed for greater accuracy in the preparation of restorations and prosthetic components [5–8]. These systems are divided into the extraoral (indirect) and the intraoral (direct) techniques [9]. Recent studies have verified a similar or better precision of the digital scans as that obtained by conventional
⁎ Corresponding author at: Department of Dental Materials and Prosthodontics, UNESP - São Paulo State University, José Bonifácio st., 1193, Vila Mendonça, Araçatuba, São Paulo 16050-050, Brazil. E-mail addresses: [email protected], [email protected] (E.P. Pellizzer).
https://doi.org/10.1016/j.msec.2019.03.059 Received 8 September 2018; Received in revised form 18 March 2019; Accepted 18 March 2019 Available online 19 March 2019 0928-4931/ © 2019 Published by Elsevier B.V.
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
suitable digital data from G1, a contrast powder (CEREC Optispray, Sirona) was applied under the resin matrix surface to be scanned. Ten other scans of the previous plaster models were made using the extraoral system with the aid of the same transfers used in the anterior step (Fig. 1). The data from Groups 2 and 3 were obtained in standard tessellation language (.STL) format and transferred to the Implant-Abutment Module software program (DWOS – Dental Wings) to design the G1, G2, and G3 implant-supported frameworks. The frameworks had standard thicknesses of 0.6 mm for the crown and 4 mm for the connectors, comprising multiple implants and without an anti-rotational system. The obtained images were displayed at different angles to observe any flaws and then transferred to milling on the Industrial Complex (Conexão Sistemas de Prótese) for the open groups (G2 and G3). For Group I, which is a closed system, the digital data were obtained as .inlab file formats and were milled in a machine from the same manufacturer (InLab MC XL, Sirona, Bensheim, Hessen, Germany) and using the same material used in the other groups (pre-sintered zirconia - VITA Zahnfabrik, Bad Säckingen, Germany). All groups in the CAD-CAM systems were milled in pre-sintered zirconia (VITA Zahnfabrik) held in a furnace (inFire HTC Speed, Sirona) at 1500 °C for 2 h and 20 min after the set identification of the ceramic material. For the manufacturing of the frameworks by the conventional method of lost-wax casting (G4, n = 10), 20 universal cast long abutments (UCLA) (Conexão Sistemas de Prótese Ltda.) with a pre-machined cylinder overcasting with cobalt-chromium (CoeCr) alloy were processed by the same technician under the same conditions. All the finished frameworks were taken for analysis of the vertical and horizontal marginal misfit. The 40 frameworks were screwed in the implant with a calibrated torque using a torquemeter (Conexão Sistemas de Prótese Ltda) under the resin matrix. The VM, PHM and NHM measurements were performed in a 3D optical microscope (Fig. 2) with a digital table, 350× magnification, and a precision of 1 μm. The measures were calculated using the QSPAK computer program (Mitutoyo). Nine equidistant and predetermined points were demarcated 1 mm below the implant platform in order to standardize the sequence and position the measurements. A total of 1440 measurements were performed and calculated as follow: 2 dental elements × (9 vertical points × 40 samples) + 2 dental elements × (9 horizontal points × 40 samples). For the analysis of the results, initially, a first examiner was responsible for the numbering of the 40 frameworks. Then, a second examiner performed the randomization of 40 numbers using Research Randomizer (Version 4.0). A single operator was responsible for all the measurements of marginal misfit and the data were obtained in micrometers. Twenty percent of the samples (n = 8) were re-read for an intra-examiner calculation to verify the calibration of the examiner and concordance between the analyses. With respect to statistical analyses, the experimental variables or between-subject factors were as follow: manufacturing technique (CADCAM systems x conventional technique), scanning system (intraoral system x extraoral system) and the use of the data acquisition (contrast powder x without contrast powder). The values of the measurements were submitted for statistical analysis using the SigmaPlot software program (SigmaPlot 13, Systat Software). The level of significance was 5%.
impression methods [10,11]. However, although different CAD-CAM systems are available for use, little is known about the most accurate scanning and production technologies used in the manufacturing of implant-supported components and restorations. These systems are being widely evaluated using a variety of methodologies, and studies have demonstrated a small value of marginal misfit with the use of the CAD-CAM technology [12,13]. However, studies on the correlation or real consequence that a prosthetic misfit can cause on the components involved in the implant-prosthesis components are scarce. Different methodologies are used to verify the biomechanical effects of prosthetic misfit [14–16]. The three-dimensional finite element analysis (3D-FEA) has been shown to be an effective tool for analyzing the biomechanical behaviors of prostheses and surrounding structures and has been used to evaluate the effect of a misfit on the biomechanical behavior of implant-supported prostheses [17–19]. Research on the scientific basis of the treatment or rehabilitation with implant-supported prosthetic restorations and components made by the CAD-CAM systems and stress distribution generated on the prosthetic components are of great importance. Thus, the aim of this study was to measure the vertical misfit (VM), positive horizontal misfit (PHM), and negative horizontal misfit (NHM) of frameworks made by different CAD/CAM methods and to compare them with the frameworks manufactured by the conventional method (lost-wax casting technique). In addition, the misfit results obtaining from each group was simulated in 3D-FEA to verify the misfit influence on stress distribution in the prosthetic components. The null hypotheses formulated were as follows: (1) the misfit frameworks do not present difference in the values of marginal misfit, independent of the manufacturing method used; and (2) the misfit implant-supported prostheses have no influence in the stress distribution. 2. Materials and methods 2.1. Analysis of vertical and horizontal misfit Four groups were obtained for VM and HM analyses, according to Table 1. Two external hexagon implants (Conexão Sistemas de Prótese) with a prosthetic platform of 4.1 mm in diameter and 10.0 mm in height were embedded in acrylic resin (JET, Clássico) for making a standard matrix and simulating a three-element, implant-supported prosthesis in the mandibular posterior region (1st premolar, 2nd premolar [pontic], and 1st molar) (Fig. 1). Twenty conventional matrix resin impressions were made by double mixture using polyvinyl siloxane (Elite H-D Putty Soft Normal Setting Zhermack) with an open technique with squared transfer copings, giving rise to 20 plaster models with die stone type IV (Elite Rock Thixotropic – Zhermack) to fabricate G3 (extraoral scanner, 3Series, DWOS Dental Wings) and G4 (lost-wax casting technique) frameworks (Fig. 1). Three different digital impression systems were used to obtain the digital data of the implants (Table 2). Twenty scans were performed using the resin matrix with intraoral systems (n = 10) and two specific transfers (Scan Body Unico – Simbiosi) for use with the CAD-CAM systems (G1: Cerec Bluecam, Sirona; and G2: iTero, Cadent), according to the guidelines recommended by each manufacturer. To obtain the Table 1 Experimental groups. Groups (n = 10)
Manufacturing systems
G1
Cerec BlueCam/Sirona (intraoral closed system + contrast powder) iTero/Cadent (intraoral open system without contrast powder) 3Series/DWOS (extraoral scanning system) Conventional method of lost-wax casting (control)
G2 G3 G4
2.2. Three-dimensional finite element analysis (3D-FEA) This methodology was performed following protocols from previous studies [20–22]. The mean of marginal misfit data measured in the previous methodology were calculated according to each point measured throughout the perimeter of each framework in each group. These mean of misfit data were used to simulation of the 3D-FEA misfit model for each group (G1, G2, G3 and G4) to evaluate biomechanical behavior 67
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 1. Illustrative image of the matrix and the implants that were used and all steps for frameworks fabrication of each group. Nine demarcated points around the implant platform to standardize the measurement regions of the misfit.
of implants, screw fixation and frameworks. According to Table 3, six 3D models (one for each group, and two control without marginal misfit and four models with marginal misfit data mentioned above) were simulated. Each model represent a bone block section (cortical and trabecular), with two external hexagon implants (4.0 × 10 mm), supporting a three-unit fixed prostheses (1st premolar, 2nd premolar [pontic] and 1st molar) with metal-ceramic or metal-free screwed crowns. The bone block was obtained from a computed tomography image (in .dicom format) using Invesalius 3.0.0., (CTI Renato Archer) simulating a bone with a layer of cortical bone thickness of 1 mm surrounding a nucleus of low-density trabecular bone [23]. The implant design and prosthetic components (UCLA abutment) were obtained from original drawings. The crowns were obtained through a surface scanning of a dental mannequin using a 3D scanner (Scanner 3D MDX20, Roland DG). All structures were exported to the Rhinoceros 3D 4.0. program (NURBS Modeling for Windows), for simplification of solids, but maintained the same characteristics as the originals. The whole set (implant/abutment/crown) was incorporated into the previously modeled bone block for further export of the solids to the finite element program [Fig. 3A–G]. After the modeling phase, all geometries were imported into the Femap 11.2 (Siemens PLM Software Inc., Santa Ana, CA, USA) for preand post-processing. The FEMAP was used to create finite element models from meshes of tetrahedral parabolic solid elements [Fig. 3H,
J]. All mechanical properties of each simulated material were attributed to the meshes according to previously published studies [17,24–26] (Table 4). For this analysis, all the materials involved in the study were considered homogeneous, isotropic, and linear. Furthermore, symmetric welds were simulated for all contacts, except for the contact between implant-abutment connections, which was simulated by symmetric contact. Constraint conditions were fixed in all axes (x, y, and z) at the boneblock sections (anterior and posterior surfaces), while all other parts in the models were unrestricted. The applied force were a total of 400 N in the axial direction (points determined on the internal slope of each cusps, totaling 50 N for cusps) and 200 N in the oblique direction (45° in relation to the long axis of the implant, only for internal slope of the buccal cusps) [27–30] (Fig. 3I). After the final elaboration of the finite element models, load incorporation, and movement restriction, they were exported to NEi Nastran 11 program (Nei Software) to analyze the previously modeled mathematical problem. After resolving the mathematical calculations in the NEi Nastran 11 program (Nei Software, Westminster, California, USA), the models were reimported into the Femap to visualize the results using the von Mises stresses (vM) maps with Megapascal (MPa) units.
Table 2 Selected characteristics of the scanning systems used in the present study, as provided by the respective manufacturer. Token
System
Manufacturer
Measurement principle
Measurement type
Measurement prerequisites
Output format
CEREC
CEREC Ac Bluecam
Sirona Dental Systems
Active triangulation
Powdering
.inlab
ITE
iTero
Cadent
Parallel confocal imaging
–
.stl
3S
3Series
Dental Wings Open Software
Laser triangulation to three axis of freedom
Still images. Accuracy of 15 μm Still images. Accuracy of 15 μm Still images. Accuracy < 20 μm
–
.stl
68
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 2. (A) Matrix with the demarcated measurement points on each implant. (B) Device/matrix/framework set under the microscope's digital table for further analysis of misfits. (C) Illustrative image of the computer program connected to the microscope, which were used for the measurements. (D) Example of a vertical marginal misfit region. (E) Example of a horizontal marginal misfit region.
Table 3 Models specifications. Models
Description
C1 C2 G1 G2 G3 G4
CONTROL - Completed fit framework manufacturing with Y-TZP (Yttria Tetragonal Zirconia Polycrystal). CONTROL - Completed fit framework manufacturing with NieCr (nickel chromium alloy). Misfit framework manufacturing with Y-TZP based on the misfit values of the G1 group frameworks (Cerec Bluecam CAD-CAM system - Sirona). Misfit framework manufacturing with Y-TZP based on the misfit values of the G2 group frameworks (iTero CAD-CAM system - Cadent). Misfit framework manufacturing with Y-TZP based on the misfit values of the G3 group frameworks (3Series CAD-CAM system - DWOS). Misfit framework manufacturing with NieCr based on the misfit values of the G4 group frameworks (Conventional Method of lost-wax casting).
Fig. 3. Schematic view of simulated model (A: final model; B: crowns; C: fixation screw; D: frameworks; E: dental implants; F: cortical bone; G: trabecular bone), meshes (H: lateral view; J: occlusal view); loading applied (I). Table 4 Mechanical properties of simulated materials. Structure
Elastic modulus (E) (GPa)
Poisson coefficient (ν)
References
Low-density trabecular bone (type III bone) Cortical bone Titanium (abutment, implant) Ni-Cr alloy Zirconia Feldspathic porcelain
1,37 13,7 110,0 200,0 205,0 82,8
0,30 0,30 0,35 0,33 0,22 0,35
Sevimay et al. [38] Sertgoz. [39] Sertgoz. [39] Pierrisnard et al. [40] Coelho et al. [41] Papavasiliou et al. [42]
3. Results
statistically more accurate (p < 0.001) than the CAD-CAM systems (Fig. 4A). In a specific analysis, the G1 (Cerec Bluecam) group had the highest mean value of VM at 41.77 μm, followed by 21.01 μm for G2 (iTero), 20.71 μm for G3 (3Series), and 16.73 μm for G4 (conventional) (Fig. 4B). When comparing the G1 with the G2, it was observed that the mean G1 values were statistically higher than the G2 values (p < 0.05) (Fig. 4B). In the comparative analysis between G2 and G3, it was observed that there was no significant difference between the groups (p = 0.319), in contrast to what was observed in the comparative analysis between G3 and the G1 (p < 0.001) (Fig. 4B). When comparing the PHM values of the CAD-CAM systems versus the conventional method, it was observed that the conventional method
3.1. Marginal misfit analysis Based on the systematic error, the intra-examiner analysis (before and after values of 20% of the frameworks, n = 8), indicated that there was no significant difference in the before and after analysis (VM = paired t-test: 1.563; p = 0.120 e HM = paired t-test: 0.829; p = 0.408). The random error coefficient (Dahlberg) indicated error margins of 10 μm and 2 μm for the values of vertical and horizontal marginal misfit, respectively. For the VM, the comparison of the CAD-CAM systems versus the conventional method showed that the conventional method was
69
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Table 5 Mean values of vertical marginal misfit (μm), of each point in each group, obtained by optical microscopy and used in the generation of finite element models. Fig. 4. Graphs of comparative analysis (A: CAD-CAM systems and conventional method for VMD; B: Difference between the groups evaluated for VMD; C: Analysis for NHM of CAD).
Points
1 2 3 4 5 6 7 8 9
Group 1
Group 2
Group 3
Group 4
PM
M
PM
M
PM
M
PM
M
44 43 32 37 30 38 44 42 35
57 60 51 36 36 36 45 42 43
46 32 18 12 9 19 24 24 38
12 17 23 18 22 21 22 13 9
24 25 14 16 11 23 25 25 13
16 31 28 23 26 22 23 15 12
22 15 18 25 26 32 25 16 14
11 14 14 6 7 13 16 17 12
was more accurate than the CAD-CAM systems, and it did not present any misfit related to this subject (over-contour). In the comparative analysis between the different intraoral scanning systems G1 and G2, no significant difference was observed between the groups (G1: 128.750 μm; G2 = 126.46 μm) (p = 0.734); however, when compared (G2 = 126.46 μm) with (G3: 116.79 μm) significant difference was observed (p = 0.018). For the NHM, the CAD-CAM systems (general average: −61.91 μm) were statistically more accurate than the conventional method (G4: −95.36 μm) (p = 0.014) (Fig. 4C). Among the CAD-CAM systems evaluated, the most accurate system was G2, which did not present NHM values, followed by G3, which presented only two values of NHM.
3.2. Finite element analysis The values of marginal misfit (μm) for each measured point obtained by in vitro study were used to simulate the 3D models (Table 5). Under axial loading, it is observed that the frameworks material did not influence the general stress distribution pattern of the misfit models. In the individual analysis of the frameworks, seen through an internal view, it can be observed that the completed fit models presented a uniform stress distribution with low stress values (6 to 75 MPa) in the cervical region of each framework. However, the models with misfit values present a high stress in the region of misfit (16 to 200 MPa) with low stress concentration in the G1 (highest VM) closer to the stress distribution of the control groups (Fig. 5). In the sagittal view of the prosthetic components, it was observed that the control groups models (C1 and C2) presented similar stress distribution with values lower than 75 MPa, whereas they showed higher stress concentrations among the misfit models, especially in the screw fixation of first molar (approximately 175 MPa). The Cerec Bluecam (G1, closed intraoral system with contrast powder) system model demonstrated a higher stress concentration in the fixation screw, followed by the 3Series (G3, open extraoral and powder free), iTero (G2, open intraoral and without contrast powder), and conventional methods (G4) (Figs. 6 and 7). Under oblique loading, all the models showed higher stress concentrations in the oblique loading when compared with the axial loading. The control groups were slightly more favorable than the misfit models, which presented greater stress at the lingual face of the cervical area of frameworks (Fig. 7). The Cerec Bluecam scanning system (G1, intraoral, closed, and with contrast powder) showed a slightly higher stress intensity and concentration than the other misfit models (Fig. 7). In the analysis of prosthetic components in sagittal view, the stress distribution pattern of all the models was similar to each other, with higher intensities of stress (150 to 450 MPa) detected in the fixation screws (Figs. 7 and 8).
70
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 5. von Mises stress maps: internal view of the frameworks. Axial and oblique loading. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
Fig. 6. von Mises stress maps: sagittal cut (anteroposterior), lingual view of prosthetic structures. Axial load. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
4. Discussion
the lowest VM values when compared with the misfit of the frameworks made by the CAD-CAM systems. Although many studies have observed high levels of accuracy of the CAD-CAM systems in implantology [5–8], the superiority of the conventional method determined in this study may be related to the fact that pillars with metal bases were used in lost-wax casting, which has its base mechanically made using computer numerical control in a similar way that the dental implants are made. Thus, it optimizes the
The creation of implant-supported dentures through CAD-CAM systems has become a practical and highly predictable practice for clinicians. The association of the two methodologies carried out in this study demonstrated that the first null hypothesis formulated was rejected because it was observed through optical microscopic measurements that the prostheses made by the conventional method obtained 71
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
Fig. 7. von Mises stress maps: vestibular view of individualized prosthetic screws. Axial and oblique loading. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
Fig. 8. von Mises stress maps: sagittal cut (anteroposterior), lingual view of prosthetic structures. Oblique loading. C1: completed fit zirconia framework, C2: completed fit metallic framework, G1: misfit zirconia framework - Cerec Bluecam, G2: misfit zirconia framework - iTero, G3: misfit zirconia framework - 3Series, G4: misfit metal framework - conventional method.
of frameworks made by an intraoral scanning system (G2 [iTero]) and an extraoral scanning system (G3 [3Series]) (p = 0.319), which were also very close to the values obtained with the use of the conventional method (G4). At this point the literature is still not very clear, as some authors [32] believe that intraoral scanning is capable of offering greater precision, whereas others point out that the accuracy of extraoral scanning systems may be directly related to the technology used in
accuracy of adaptation between the abutment and platform of the implant, minimizing the misfits between components. Moreover, as observed in the study by Kahramaoglu & Kulak-Ozkan [31], the different values of the thermal expansion coefficient of zirconia, metallic alloy of the prosthetic component (CreCo), and metallic infrastructure nickelchromium (NieCr) may have influenced the results of this study. In our study, similar values of VM were observed in the comparison 72
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
movements [49], in addition to all the clinical procedures, have been performed under ideal laboratory conditions. The results observed by FEA analysis permit the understanding of biomechanical behavior of different structures; however, the results should be interpreted carefully due to limitation of computation simulation, such as materials properties are assumed to be isotropic, homogeneous, linear and static occlusal loading [16,20]. So, these limitations must be kept in mind before of extrapolated to the clinic day. However, when combined with trial clinical studies, these biomechanical analyses could serve as a guideline for use of the clinicians in implant dentistry. Therefore, further randomized-controlled trials should be conducted to gain a better understanding of the influence of marginal misfit in long term follow-up o implant-supported prostheses.
the scanning camera, as well as the optical quality of the pulverized material, such as contrast powder when used, and the resolution of the surface scanning device [33,34]. When comparing the two intraoral scanning systems used in this study, we observed that the system requiring the application of a contrast powder had the highest mean VM values (p < 0.05). Although the application of the powder facilitates the surface scanning, it can introduce a certain imprecision of scanning, due to the variation of its thickness (5 μm to 15 μm) [35]. However, it should be noted that all VM values reported in this study are within the clinically acceptable limits (i.e., 100 to 150 μm), to avoid mechanical and biological complications [36–38]. With regards to the PHM, the conventional method was more accurate and did not present frameworks with overextension or over configuration, whereas the CAD-CAM systems had lower average values of NHM than the conventional method. The frameworks of pre-sintered zirconia, on average, are milled 20% higher than the planned final dimensions. Before starting the milling procedures, a bar code reading in each ceramic block is performed, in which the milling software identifies the contraction rate of each material after being submitted to the sintering process [39–41]. For this reason, the results determined in this study may be related to this dimensional alteration encountered in the zirconia frameworks during the sintering, or to the fact that manual adjustments were not made in any part of this study. The second null hypothesis formulated in this study was also rejected. Using 3D-FEA, we evaluated the biomechanical behavior of the implant-prosthesis set, and we observed that prosthetic misfit influenced the stress distribution of the components evaluated. However, in this study, the frameworks of manufacturing material and small variations of misfit between the groups did not change the distribution pattern and the concentration of stress, corroborating previous studies [14,42,43]. There were no significant differences between the models with oblique loading. However, with axial loading, the conventional method (G1) presented the best biomechanical situation, which presented the lowest marginal misfit values. Vertical misfits are mainly related to, among other factors, fractures of the prosthetic components or frameworks, abutment and/or screw loosening. Misfits are associated with increased difficulty in achieving satisfactory dental hygiene. Consequently, they may induce greater retention of food and cause a greater bacterial accumulation [44], altering the buccal microflora and increasing the possibilities of the development of inflammatory reactions and infections in the peri-implant region, and may even increase the rate of marginal bone loss [42,45]. Although research studies strive to determine the clinically acceptable values of marginal misfits, little is known about the borderline misfit that does not significantly impair bone tissue and osseointegration of the implant. Thus, understanding the difficulty involved in obtaining an absolute passive settlement of the implant-supported components and restorations, most authors recommend minimizing, as much as possible, the misfits to avoid the development of the previously-mentioned possible complications [45–47]. Another recommendation is to establish a minimum passive settling of the components, in which the formation of static charges and high intensity stresses inside the prosthesis or in the surrounding bone tissue will not be induced [45,48]. Finally, from this study, we can consider that although the conventional method presented the least values of marginal misfit, both the intraoral and extraoral scanning technique can be predicted in the manufacturing of implant-supported prostheses, presenting frameworks with the best possible adaptation and optimizing the longevity of rehabilitation treatments. However, the accelerated development and improvement of the CAD-CAM systems in dentistry must be accompanied by constant professional training [50]. The limitations to this study are inherent to the in vitro methodologies that were used. The preparation of a matrix differs from a natural intraoral scan situation involving soft tissue, saliva, and patient
5. Conclusions Based on the methodologies and within the limitations of this in vitro study, the following conclusions were drawn: 1. The manufacturing method used influenced the misfit of the frameworks, for both vertical and horizontal measurements. With the lowest means of misfit, the conventional method being the most favorable and the Cerec Bluecam System the least favorable; 2. The frameworks misfit influenced the stress distribution in the prosthetic components, while the completely fit prostheses were more biomechanically favorable than the misfit prostheses. Acknowledgments Supported by (Scholarship) CNPq: 165406/2015-1 and São Paulo Research Foundation (FAPESP) Grant: #2009/16164-7, #2011/071347 and #2011/19150-7. References [1] E.A. Patterson, R.B. Johns, Theoretical analysis of the fatigue life of fixture screws in osseointegrated dental implants, Int. J. Oral Maxillofac. Implants 7 (1) (1992 Spring) 26–33. [2] W.G. Assunção, E.A. Gomes, E.P. Rocha, J.A. Delben, Three-dimensional finite element analysis of vertical and angular misfit in implant-supported fixed prostheses, Int. J. Oral Maxillofac. Implants 26 (2011) 788–796. [3] G. Priest, Virtual-designed and computer-milled implant abutments, J. Oral Maxillofac. Surg. 63 (2005) 22–32. [4] A. Syrek, G. Reich, D. Ranftl, C. Klein, B. Cerny, J. Brodesser, Clinical evaluation of all-ceramic crowns fabricated from intraoral digital impressions based on the principle of active wavefront sampling, J. Dent. 38 (2010) 553–559. [5] P. Vigolo, F. Fonzi, Z. Majzoub, G. Cordioli, An in vitro evaluation of titanium, zirconia, and alumina procera abutments with hexagonal connection, Int. J. Oral Maxillofac. Implants 21 (2006) 575–580. [6] P. Coli, S. Karlsson, Precision of a CAD-CAM technique for the production of zirconium dioxide copings, Int. J. Prosthodont. 17 (2004) 577–580. [7] I. Naert, A. Van der Donck, L. Beckers, Precision of fit and clinical evaluation of allceramic full restorations followed between 0.5 and 5 years, J. Oral Rehabil. 32 (2005) 51–57. [8] D.G. de França, M.H. Morais, F.D. das Neves, A.F. Carreiro, G.A. Barbosa, Precision fit of screw-retained implant-supported fixed dental prostheses fabricated by CADCAM, copy-milling, and conventional methods, Int. J. Oral Maxillofac. Implants (2016), https://doi.org/10.11607/jomi.5023 (Epub ahead of print). [9] H. Rudolph, H. Salmen, M. Moldan, K. Kuhn, V. Sichwardt, B. Wöstmann, et al., Accuracy of intraoral and extraoral digital data acquisition for dental restorations, J. Appl. Oral Sci. 24 (2016) 85–94. [10] F.S. Andriessen, D.R. Rijkens, W.J. van der Meer, D.W. Wismeijer, Applicability and accuracy of an intraoral scanner for scanning multiple implants in edentulous mandibles: a pilot study, J. Prosthet. Dent. 111 (2014) 186–194. [11] W.J. van der Meer, F.S. Andriessen, D. Wismeijer, Y. Ren, Application of intra-oral dental scanners in the digital workflow of implantology, PLoS One 7 (2012) e43312. [12] M.F. Leifert, M.M. Leifert, S.S. Efstratiadis, T.J. Cangialosi, Comparasion of space analysis evaluations with digital models and plaster dental casts, Am. J. Orthod. Dentofac. Orthop. 136 (2009) 16.e1–16.e4. [13] H.A. Mously, M. Finkkelman, R. Zandparsa, H. Hirayama, Marginal and internal adaptation of ceramic crown restorations fabricated with CAD-CAM technology and the heat-press technique, J. Prosthet. Dent. 112 (2014) 249–256. [14] J. Abduo, K. Lyons, Effect of vertical misfit on strain within screw-retained implant titanium and zirconia frameworks, J. Prosthodont. Res. 56 (2012) 102–109.
73
Materials Science & Engineering C 102 (2019) 66–74
C.C. Mello, et al.
[15] D.K. de Vasconcellos, A.N. Kojima, A.M. Mesquita, M.A. Bottino, M. Ozcan, A microstrain comparison of passively fitting screw-retained and cemented titanium frameworks, J. Prosthet. Dent. 112 (4) (2014) 834–838. [16] F.R. Verri, J.F. Santiago Jr., D.A. Almeida, V.E. de Souza Batista, C.A. Araujo Lemos, C.C. Mello, et al., Biomechanical three-dimensional finite element analysis of single implant-supported prostheses in the anterior maxilla, with different surgical techniques and implant types, Int. J. Oral Maxillofac. Implants 32 (2017) e191–e198. [17] J. Abduo, R.B. Judge, Implications of implant framework misfit: a systematic review of biomechanical sequelae, Int. J. Oral Maxillofac. Implants 29 (2014) 608–621. [18] C. Manzella, C. Bignardi, V. Burello, S. Carossa, G. Schierano, Method to improve passive fit of frameworks on implant-supported prostheses: an in vitro study, J. Prosthet. Dent. 116 (2016) 52–58. [19] M.B. dos Santos, A. Bacchi, R.L. Consani, L. Correr-Sobrinho, Influence of diamondlike carbon coating of screws on axial tightening force and stress distribution on overdenture bar frameworks with different fit levels and materials, Int. J. Oral Maxillofac. Implants 30 (5) (2015 Sep-Oct) 1019–1027 (30). [20] E.P. Pellizzer, C.A.A. Lemos, D.A.F. Almeida, V.E. de Souza Batista, J.F. Santiago Júnior, F.R. Verri, Biomechanical analysis of different implant-abutments interfaces in different bone types: an in silico analysis, Mater. Sci. Eng. C Mater. Biol. Appl. 90 (2018) 645–650 Sep 1. [21] L. Minatel, F.R. Verri, G.A.H. Kudo, D.A. de Faria Almeida, V.E. de Souza Batista, C.A.A. Lemos, E.P. Pellizzer, J.F. Santiago, Junior. Effect of different types of prosthetic platforms on stress-distribution in dental implant-supported prostheses, Mater. Sci. Eng. C Mater. Biol. Appl. 71 (2017) 35–42 Feb 1. [22] J.F. Santiago Junior, F.R. Verri, D.A. Almeida, V.E. de Souza Batista, C.A. Lemos, E.P. Pellizzer, Finite element analysis on influence of implant surface treatments, connection and bone types, Mater. Sci. Eng. C Mater. Biol. Appl. 63 (2016) 292–300 Jun. [23] U. Lekholm, G.A. Zarb, P.-I. Brånemark, G.A. Zarb, T. Albrektsson, Tissue-integrated prostheses, Tissue Integrated Prostheses, Quintessence, Chicago, 1985, pp. 199–209. [24] M. Sevimay, F. Turhan, M.A. Kiliçarslan, G. Eskitascioglu, Three-dimensional finite element analysis of the effect of different bone quality on stress distribution in an implant-supported crown, J. Prosthet. Dent. 93 (2005) 227–234. [25] A. Sertgoz, Finite element analysis study of the effect of superstructure material on stress distribution in an implant supported fixed prosthesis, Int. J. Prosthodont. 10 (1997) 19–27. [26] P.G. Coelho, E.A. Bonfante, N.R. Silva, E.D. Rekow, V.P. Thompson, Laboratory simulation of Y-TZP all-ceramic crown clinical failures, J. Dent. Res. 88 (2009) 382–386. [27] C.A.A. Lemos, F.R. Verri, J.F. Santiago Junior, V.E. de Souza Batista, D.T. Kemmoku, P.Y. Noritomi, E.P. Pellizzer, Splinted and nonsplinted crowns with different implant lengths in the posterior maxilla by three-dimensional finite element analysis, J. Healthc. Eng. 2018 (2018 Sep 3) 3163096, https://doi.org/10. 1155/2018/3163096. [28] V.E. de Souza Batista, F.R. Verri, D.A.F. Almeida, J.F. Santiago Junior, C.A.A. Lemos, E.P. Pellizzer, Evaluation of the effect of an offset implant configuration in the posterior maxilla with external hexagon implant platform: a 3-dimensional finite element analysis, J. Prosthet. Dent. 118 (3) (2017) 363–371 Sep. [29] C.A.A. Lemos, F.R. Verri, Santiago J.F. Júnior, D.A.F. Almeida, V.E.S. Batista, P.Y. Noritomi, E.P. Pellizzer, Retention system and splinting on Morse taper implants in the posterior maxilla by 3D finite element analysis, Braz. Dent. J. 29 (1) (2018) 30–35 Jan-Feb. [30] V.E. de Souza Batista, F.R. Verri, D.A. Almeida, J.F. Santiago Junior, C.A. Lemos, E.P. Pellizzer, Finite element analysis of implant-supported prosthesis with pontic and cantilever in the posterior maxilla, Comput. Meth. Biomech. Biomed. Eng. 20 (6) (2017) 663–670 May. [31] E. Kahramanoğlu, Y. Kulak-Ozkan, Marginal and internal adaptation of different superstructure and abutment materials using two different implant systems for five-
[32]
[33] [34] [35]
[36]
[37] [38]
[39] [40]
[41]
[42]
[43]
[44]
[45]
[46] [47]
[48]
[49]
[50]
74
unit implant-supported fixed partial dentures: an in vitro study, Int. J. Oral Maxillofac. Implants 28 (2013) 1207–1216, https://doi.org/10.11607/jomi.2744. T.V. Flügge, S. Schlager, K. Nelson, S. Nahles, M.C. Metzger, Precision of intraoral digital dental impressions with iTero and extraoral digitization with the iTero and a model scanner, Am. J. Orthod. Dentofac. Orthop. 144 (2013) 471–478, https://doi. org/10.1016/j.ajodo.2013.04.017. M. Persson, M. Andersson, B. Bergman, The accuracy of a high-precision digitizer for CAD-CAM of crowns, J. Prosthet. Dent. 74 (1995) 223–229. H.H. Zaghloul, J.F. Younis, Marginal fit of implant-supported all-ceramic zirconia frameworks, J. Oral Implantol. 39 (2013) 417–424. J. Katsoulis, R. Mericske-Stern, D.M. Yates, N. Izutani, N. Enkling, M.B. Blatz, In vitro precision of fit of computer-aided design and computer-aided manufacturing titanium and zirconium dioxide bars, Dent. Mater. 29 (2013) 945–953. T. Jemt, Failures and complications in 391 consecutively inserted fixed prostheses supported by Brånemark implants in edentulous jaws: a study of treatment from the time of prosthesis placement to the first annual checkup, Int. J. Oral Maxillofac. Implants 6 (1991) 270–276. T. Jemt, J.E. Rubenstein, L. Carlsson, B.R. Lang, Measuring fit at the implant prosthodontic interface, J. Prosthet. Dent. 75 (1996) 314–325. E. Bayramoğlu, Y.K. Özkan, C. Yildiz, Comparison of marginal and internal fit of press-on-metal and conventional ceramic systems for three- and four-unit implantsupported partial fixed dental prostheses: an in vitro study, J. Prosthet. Dent. 114 (2015) 52–58. F. Komine, M.B. Blatz, H. Matsumura, Current status of zirconia-based fixed restorations, J. Oral Sci. 52 (2010) 531–539. F. Filser, P. Kocher, F. Weibel, H. Lüthy, P. Schärer, L.J. Gauckler, Reliability and strength of all-ceramic dental restorations made by direct ceramic machining (DCM), Int. J. Comput. Dent. 4 (2001) 89–106. D. Suttor, K. Bunke, S. Hoescheler, H. Hauptmann, G. Hertlein, LAVA - the system for all-ceramic ZrO2 crown and bridge frameworks, Int. J. Comput. Dent. 4 (2001) 195–206. C. Kunavisarut, L.A. Lang, B.R. Stoner, D.A. Felton, Finite element analysis on dental implant-supported prostheses without passive fit, J. Prosthodont. 11 (2002) 30–40. R.T. Abreu, A.O. Spazzin, P.Y. Noritomi, R.L. Consani, M.F. Mesquita, Influence of material of overdenture-retaining bar with vertical misfit on three-dimensional stress distribution, J. Prosthodont. 19 (2010) 425–431. F.D. Neves, C.J. Prado, M.S. Prudente, T.A. Carneiro, K. Zancopé, L.R. Davi, et al., Micro-computed tomography evaluation of marginal fit of lithium disilicate crowns fabricated by using chairside CAD-CAM systems or the heat-pressing technique, J. Prosthet. Dent. 112 (2014) 1134–1140. M.M. Buzayan, N.B. Yunus, Passive fit in screw retained multi-unit implant prosthesis understanding and achieving: a review of the literature, J. Indian Prosthodont. Soc. 14 (2014) 16–23. S. Sahin, M.C. Cehreli, The significance of passive framework fit in implant prosthodontics: current status, Implant. Dent. 10 (2001) 85–92. S.A. de Sousa, M.A. de Arruda Nobilo, G.E. Henriques, M.F. Mesquita, Passive fit of frameworks in titanium and palladium-silver alloy submitted the laser welding, J. Oral Rehabil. 35 (2008) 123–127. R. Sorrentino, E.F. Gherlone, G. Calesini, F. Zarone, Effect of implant angulation, connection length, and impression material on the dimensional accuracy of implant impressions: an in vitro comparative study, Clin. Implant. Dent. Relat. Res. 12 (Suppl. 1) (2010) e63–e76. S.Y. Kim, M.J. Kim, J.S. Han, I.S. Yeo, Y.J. Lim, H.B. Kwon, Accuracy of dies captured by an intraoral digital impression system using parallel confocal imaging, Int. J. Prosthodont. 26 (2013) 161–163. G.Á. Galhano, E.P. Pellizzer, J.V. Mazaro, Optical impression systems for CAD-CAM restorations, J. Craniofac. Surg. 23 (2012) e575–e579.