Influence of Abutment Design on Stiffness, Strength, and Failure of Implant-Supported Monolithic Resin Nano Ceramic (RNC) Crowns Tim Joda, Dr Med Dent, MSc;*,† Samuel Huber, Med Dent;* Alexander Bürki, MSc;‡ Philippe Zysset, PhD;‡ Urs Brägger, Dr Med Dent*
ABSTRACT Background: Recent technical development allows the digital manufacturing of monolithic reconstructions with highperformance materials. For implant-supported crowns, the fixation requires an abutment design onto which the reconstruction can be bonded. Purpose: The aim of this laboratory investigation was to analyze stiffness, strength, and failure modes of implant-supported, computer-assisted design and computer-aided manufacturing (CAD/CAM)-generated resin nano ceramic (RNC) crowns bonded to three different titanium abutments. Materials and Methods: Eighteen monolithic RNC crowns were produced and loaded in a universal testing machine under quasi-static condition according to DIN ISO 14801. With regard to the type of titanium abutment, three groups were defined: (1) prefabricated cementable standard; (2) CAD/CAM-constructed individualized; and (3) novel prefabricated bonding base. Stiffness and strength were measured and analyzed statistically with Wilcoxon rank sum test. Sections of the specimens were examined microscopically. Results: Stiffness demonstrated high stability for all specimens loaded in the physiological loading range with means and standard deviations of 1,579 1 120 N/mm (group A), 1,733 1 89 N/mm (group B), and 1,704 1 162 N/mm (group C). Mean strength of the novel prefabricated bonding base (group C) was 17% lower than of the two other groups. Plastic deformations were detectable for all implant-abutment crown connections. Conclusions: Monolithic implant crowns made of RNC seem to represent a feasible and stable prosthetic construction under laboratory testing conditions with strength higher than the average occlusal force, independent of the different abutment designs used in this investigation. KEY WORDS: abutment connection, dental implants, DIN ISO 14801, failure mode, resin nano ceramic (RNC), stiffness, strength, titanium
materials.1–3 Recently, the entire prosthetic fabrication process of implant-supported reconstructions has been introduced in a complete digital workflow even without any physical models.4 In addition, the production of the suprastructure can be simplified by the option to connect crowns to prefabricated or individually customized abutments.5 In fixed implant prosthodontics, it is of interest to transfer the developments of digital computerassisted design and computer-aided manufacturing (CAD/CAM) technology to the fabrication of reconstructions with constant quality and a reasonable costbenefit ratio.6 Therefore, the demand arises for crowns made of high-strength monolithic materials, which are produced in a streamlined workflow and bonded to
INTRODUCTION Technical development in the field of digital dental medicine has opened the opportunity for the manufacturing of reconstructions using high-performance *Division of Fixed Prosthodontics, School of Dental Medicine, University of Bern, Bern, Switzerland; †Department of Prosthetic Dentistry, Center for Dental and Oral Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany; ‡Institute for Surgical Technology and Biomechanics, University of Bern, Bern, Switzerland Corresponding Author: Dr. Tim Joda, MSc, Division of Fixed Prosthodontics, School of Dental Medicine, University of Bern, Freiburgstr. 7, Bern CH-3010, Switzerland; e-mail: [email protected] © 2014 Wiley Periodicals, Inc. DOI 10.1111/cid.12215
1
2
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
TABLE 1 Physical Characteristics and Properties of Resin Nano Ceramic (RNC) Material
Density (g/cm3) Flexural strength (MPa0
RNC
2.1 204 1 19
Manufacturer information: Lava Ultimate Restorative, 3M ESPE, Neuss, Germany.
abutment substructures for one-piece screw-retained implant reconstructions.7 The potential of monolithic materials and their use especially in the functional zone of posterior regions has been under continuous investigation.3 Despite significant improvements in material properties, the clinical outcome of all-ceramic crowns still fails to match the success rates of metal ceramic-fixed reconstructions.8,9 In tooth-supported reconstructions, the fabrication method of CAD/CAM-constructed veneering material sintered to zirconia copings or full-contour zirconia crowns seem to fulfill high-loading capacities but also initiate excessive wear at natural antagonists.1,10,11 A recently defined material class – resin nano ceramic (RNC) – composed of bonded zirconia-silica nanoparticles, clustered and embedded in a cross-linked resin matrix, has been introduced for the CAD/CAM fabrication of fixed reconstructions (Lava Ultimate Restorative, 3M ESPE, Neuss, Germany).2 In preclinical tests, RNC exhibited a comparable fracture resistance to glass ceramics and a functional balance to enamel structures with a flexural modulus in the same range as dentin (Table 1). However, at present, no valid data are available in the scientific literature related to stiffness and strength of monolithic implant-supported reconstructions with RNC. The aim of this in vitro investigation was to compare stiffness, strength, and the failure modes of trans-occlusal, screw-retained, single-unit monolithic crowns made of RNC connected to two prefabricated and one individualized titanium abutment. The null hypothesis was that no difference exists in the mean stiffness, mean strength, and failure mode between the various abutment crown groups. MATERIALS AND METHODS Experimental Groups All specimens were designed as trans-occlusal, screwretained, one-piece reconstructions. The setup consisted
of a CAD/CAM-generated monolithic RNC-crown bonded to different types of abutments on soft tissue-level implants (Tissue Level Implant SLA RN 4.1 mm × 12 mm, Institute Straumann AG, Basel, Switzerland). In contrast to a naturally tooth-like appearance, the RNC crowns were fabricated as semi-sphere reconstructions with regard to the specific guidelines of DIN ISO 14801 (Dentistry-fatigue test for endosseous implants, International Organization for Standardization 2005, Geneva, Switzerland). Overall, the study included 18 test specimens, divided into 3 groups (Figure 1): •
•
•
Group A: n = 6 RNC crowns bonded to prefabricated standard titanium abutments (synOcta cementable, Institute Straumann AG); Group B: n = 6 RNC crowns bonded to individualized CAD/CAM-manufactured titanium abutments (CARES, Institute Straumann AG); Group C: n = 6 RNC crowns bonded to novel prefabricated bonding base titanium abutments (Variobase, Institute Straumann AG).
For manufacturing of the prosthetic RNC crowns, one master abutment per group was determined. The master abutments were mounted to corresponding implant analogues and digitalized with a laboratory CAD/ CAM scanner system (CARES CS2 Scanner, Institute Straumann AG). Three-dimensional virtual simulation and construction of the crowns were performed with a software program (CARES Visual Software 8.0, Institute Straumann AG). RNC blanks were used for the fabrication of the crowns in an external milling center (Straumann CAD/CAM GmbH, Markkleeberg, Germany). After milling, the RNC reconstructions were cleaned with 95% ethanol. Then, the titanium abutment types of groups A, B, and C were prepared with a specialized primer and bonded to the monolithic RNC crowns with resin cement (Panavia F 2.0, Kuraray Noritake Dental Inc., Tokyo, Japan) under laboratory conditions. Mechanical Resistance The preparation of the specimens for mechanical testing was performed according to a previously published article.12 Briefly summarized, the implants were perpendicularly embedded into a clamping device with
Implant-Supported Monolithic RNC Crowns
3
Figure 1 Study flowchart presenting a total of 18 specimens with 3 groups of different types of titanium abutments (Institute Straumann AG, Basel, Switzerland) bonded to monolithic RNC crowns (Lava Ultimate Restorative, 3M ESPE, Neuss, Germany). RNC = resin nano ceramic.
polymethyl methacrylate (PMMA) (EuroPlex SDX, Röhm AG, Brüttisellen, Switzerland) (Figure 2). The material is characterized by a modulus of elasticity higher than 3 GPa. The position was 3.0 mm 1 0.5 mm underneath the top of the implant neck in order to mimic physiologic bone loss according to DIN ISO 14801. The prepared monolithic one-piece reconstructions were mounted with 35 Ncm onto the dental implants using a manual torque control ratchet. Afterward, the specimens were clamped at a 30° 1 2° angle to the loading direction of the testing machine. Quasi-static load was applied with 1 mm/minute (Figure 3). Forces and displacements were recorded, and stiffness was defined as the maximum slope of a moving linear regression with a bandwidth that maximizes the correlation coefficient. Strength was documented as the maximal force until a decrease of 20% of the loadbearing capacity was noted.
(A)
(B)
Moreover, the mobility of implant prosthetic components was verified. In order to visualize the characteristics of failure modes and material alterations, crosssections were prepared for examination under polarized light. Specimens of each test group were embedded in methyl methacrylate and cut through the center in the direction of the loading zone into approximately 400-μm thick ground sections using a slow-speed diamond saw (Varicut VC-50, Leco, Munich, Germany). Polishing started with P100 grit silicon carbide paper and went through four levels, finishing with P600 grit. This procedure was followed by finer surface treatment with 0.05 micron alumina paste. Statistical Analysis Descriptive statistics as mean, standard deviation, and minimum and maximum of stiffness and strength were provided for the tested specimens of synOcta, CARES,
(C)
Figure 2 Technical preparation sequence of the test specimens with a specialized application tool (A), embedded in a clamping device with polymethyl methacrylate and placement 3.0 mm 1 0.5 mm underneath the top of the implant neck (B) as well as mounted RNC reconstruction (C). RNC = resin nano ceramic.
4
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Figure 3 Illustration of the testing set up according to the guidelines of DIN ISO 14801 (1 = implant reconstruction; 2 = PMMA embedding; 3 = embedding cup; 4 = fixation device; 5 = load cell; and 6 = piston). PMMA = polymethyl metharylate.
and Variobase abutments. Calculated stiffness and strength were statistically compared with Wilcoxon rank sum test. The level of statistical significance was set at p < .05. RESULTS None of the tested specimens displayed any obvious fractures after load application. All implant-abutment crown connections remained stable and no movement could be detected. Moreover, specimens of the three tested groups did not show any visible defects macroscopically. One specimen of the CARES abutment group fractured the RNC crown. Typical load displacement curves for the implant-abutment reconstructions with synOcta, CARES, and Variobase are shown in Figure 4.
Figure 4 Typical load displacement curves for the RNC implant-abutment reconstructions with synOcta abutment (group A), CARES abutment (group B), and Variobase abutment (group C). RNC = resin nano ceramic.
Figure 5 Boxplot diagram for stiffness of synOcta abutments (group A), CARES abutments (group B), and Variobase abutments (group C) with statistically significant differences between groups A and B (p < .05) (Wilcoxon rank test corrected for multiple testing by the method of Holm).
In addition to the individual results for all specimens, the mean and standard deviations of stiffness and strength of each group are reported in Table 2. The computed stiffness ranged from 1,362 N/mm to 1,876 N/ mm, while strength ranged from 1,088 N to 1,504 N. First, a Kruskal–Wallis test was performed: one each with the variable strength and the variable stiffness. The p values for strength and stiffness showed a significant difference in the overall median in both calculations. In an additional test, each two groups were compared in order to find exact differences. The Wilcoxon rank sum test compared the median of the two variables between each group. Statistical calculations revealed that mean stiffness of the Variobase group was equivalent to the means of the two other groups, but that the mean of the SynOcta group was 9% lower than the one of the CARES group (Figure 5). Mean strength of the Variobase group was 17% lower than those of the two other groups and that difference was statistically significant (Figure 6). The Variobase bonding abutment demonstrated the most homogenous results with the narrowest ranges of stiffness and strength (Figures 5 and 6). Microscopic examination of the cross-sections obtained from groups A, B, and C presented plastic deformations of the implants and abutments including screws for all tested specimens in the direction of the load application without any cracks (Figures 7 and 8).
Implant-Supported Monolithic RNC Crowns
Figure 6 Boxplot diagram of strength of synOcta abutments (group A), CARES abutments (group B), and Variobase abutments (group C) with statistically significant differences between groups A and C (p < .05) and groups B and C (p < .05) (Wilcoxon rank test corrected for multiple testing by the method of Holm).
DISCUSSION The findings of the present in vitro investigation demonstrated consistently high stiffness and strength under quasi-static loading for all three titanium abutment configurations in combination with the bonded monolithic RNC suprastructures. However, inner-group comparisons revealed 9% lower stiffness for the CARES group and 17% lower strength for prefabricated Variobase abutments compared with prefabricated synOcta and individualized CARES abutments. Because of the use of polymethyl metharylate (PMMA) as a reproducible substitute of bone for anchoring the implant, there is
no clinically relevant target value for absolute stiffness. Nevertheless, stiffness remains an important comparative variable to evaluate the proper function of load-bearing structures such as implant-supported reconstructions. Physiologic chewing forces for a single tooth in the molar area have been reported to range from 200 N to 400 N.13,14 In this physiologic load range, the mean stiffness of the Variobase design was equivalent to those of the two other groups and confers therefore similar stability. Moreover, all specimens exhibited quasi-static strength well beyond this physiological loading range. The results of laboratory studies must be translated carefully to clinical conditions, with varying loading modes as well as large number of loading cycles. For instance, in contrast to physiological chewing forces, the selected loading protocol was quasi-static. However, the failure mode was plastic deformation of metallic parts, which is essentially strain rate independent. Accordingly, based on the presented results and the current clinical experience, it can be assumed that all evaluated specimens could withstand masticatory forces. The abutment design selected in this trial seems to have no essential clinical relevance, even though some differences were statistically significant between the groups. Sample size was calculated with n = 6 specimens per test group. The technical preparation sequence of the specimens was standardized with a specialized application tool. This procedure guaranteed exact and reproducible results for embedding and placing in the clamping device in accordance to the guidelines of DIN ISO 14801.
TABLE 2 Results Displaying Maximal Forces (N) and Stiffness (N/mm) for Each Specimen of Groups A, B, and C
#
1 2 3 4 5 6 Average SD
Group A
Group B
Group C
synOcta Ti Abutment
CARES Ti Abutment
Variobase Ti Abutment
max Force (N)
Stiffness (N/mm)
max Force (N)
Stiffness (N/mm)
max Force (N)
Stiffness (N/mm)
1,409 1,504 1,435 1,412 1,146 1,196 1,350 144
1,579 1,571 1,723 1,640 1,362 1,601 1,579 120
1,398 1,502 1,287 1,275 1,335 1,306 1,350 86
1,612 1,673 1,721 1,771 1,873 1,745 1,733 89
1,153 1,117 1,147 1,113 1,088 1,132 1,125 24
1,738 1,395 1,876 1,773 1,724 1,719 1,704 162
SD = standard deviation.
5
6
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Figure 7 Cross-sectional analysis for groups A, B, and C with plastic deformations of the implant, abutment including screw, and bonded RNC crown (not visible in the embedded acrylic material) in the direction of load application without any signs of fractures (photomicrographic examinations with original ×12.5 magnification). RNC = resin nano ceramic.
For comparison, no studies were found reporting on monolithic implant reconstructions made of RNC and their mechanical behavior in the dental literature. Two recent laboratory studies, however, investigated the fracture resistance of monolithic lithium disilicate implant crowns.15,16 At molar sites, Kim and colleagues compared strength after load application of full contour lithium
Figure 8 Cross-section group C: implant, Variobase titanium abutment including screw and bonded RNC crown with close-up views of pressure (A) and tension zones (B). RNC = resin nano ceramic.
disilicate crowns to veneered zirconia substructures that were either hand layered or heat pressed.15 All crowns and zirconia copings were CAD/CAM generated with a labside milling system (Cerec, Sirona Dental System GmbH, Bensheim, Germany) and luted to the abutment substructures using resin cement. The monolithic lithium disilicate implant-supported reconstructions demonstrated a fracture load of 3,852.1 N. Heterogeneous results were reported on the zirconia-based reconstructions with significantly higher fracture loads for heat-pressed crowns (5,229.3 N) in contrast to handlayered crowns (3,100.3 N). Because of the different study design, the results are not directly comparable with the finding of the present investigations.15 In the second in vitro investigation, CAD/CAMfabricated and pressed monolithic lithium disilicate crowns were connected to customized implant abutments made of titanium and zirconia with a design for maxillary incisor replacement. In general, mean fracture resistance was significantly higher for reconstructions with titanium than zirconia abutments (p > .0001). Implant-supported crowns fabricated with CAD/ CAM-constructed monolithic lithium disilicate crowns in combination with titanium abutments exhibited the highest fracture resistance (558.5 N). All other
Implant-Supported Monolithic RNC Crowns
abutment-crown combinations ranged between 340.3 N and 495.9 N.16 The reported results for fracture resistances in case of full-contour lithium disilicate implant crowns connected to titanium abutments were obviously lower compared with the findings of the present trial with monolithic RNC crowns, regardless of the used abutment type. The interpretation of these findings may lead to the speculation that the crowns design or the restoration material itself had a significant influence of the fracture resistance of the entire reconstruction. In addition, individualized titanium and zirconia abutments with and without prosthetic suprastructures were tested under static loading in an in vitro approach identical to this investigation.17 The results of the individualized titanium abutment group demonstrated enhanced stability of unrestored test specimens (1,249 N) compared with implant-abutment reconstructions with glass-ceramic crowns (762 N). These findings were contradictory to the results of the present trial. The measured strength for the group with individualized titanium abutments connected to glass-ceramic crowns was at least 363 N lower than the average strength for the titanium abutments bonded with RNC crowns tested in this study. Thus, these results clearly illustrate the powerful connection of the monolithic RNC reconstructions with both the prefabricated and the individualized abutment types. In the aforementioned study, a remarkably high number of fractures of the glass-ceramic crowns (60% of the titanium abutment-supported crowns) were observed; and moreover, 30% of the specimens showed a fracture of the abutment screw or even of the implant. Twenty percent of the specimens showed mobility of at least one prosthetic component. These results suggest that the weakest link is the crown. In contrast to these findings, the present investigation revealed no detectable fractures after quasi-static load application. Neither any implant prosthetic component nor the implant were cracked. Based on the elastic properties of metallic materials, plastic deformation of the implant/abutment complex was observed in all test specimens. Surprisingly, none of the tested titanium abutments or the bonding connection to the monolithic RNC crown was mobile after loading. The abutment screws were able to stabilize the implant/ abutment complex even after irreversible deformation of the implant shoulder. The implant/abutment complex, the screw connections as well as the bonding
7
techniques for the monolithic RNC crowns represented a strong mechanical unit. A possible explanation for the higher strength could be the specific physical properties of RNC, particularly in contrast to lithium disilicate or glass ceramic. The composition of bonded zirconia-silica nanoparticles in a cross-linked resin matrix had the potential to imitate a comparable flexural modulus in the same range as human dentin (Table 1). The implementation of new technologies and treatment concepts within the digital workflow enable the production of monolithic reconstructions in a CAD/CAM-based process with high-performance dental materials, such as RNC.18 Demanding laboratory work steps may therefore be streamlined and the materialspecific advantages are ensured because of standardized production quality by using industrially manufactured material blanks.4,5 For these reasons, fully anatomical reconstructions made of RNC open a broader application with an enlarged range of indications in fixed implant prosthodontics. Because of the in vitro design of the study, the present investigation had some limitations, as an artificially crown design and no consideration of thermal effects, that complicate the direct comparison and translation to clinical situations. Nevertheless, the results of quasi-static load testing provide valid information and reveal important tendencies for treatment concepts with this newly defined class of monolithic restoration material. However, subsequent trials including fatigue cycling and thermal testing should be initiated to investigate the performance and durability of implant-supported reconstructions made out of RNC in combination with the bonding agent. Moreover, further clinical studies with close follow-up observations are necessary to confirm these laboratory results. CONCLUSIONS Monolithic implant crowns made of RNC are stable prosthetic reconstructions under laboratory testing conditions with high quasi-static mechanical properties compared with physiological occlusal force. The prefabricated cementable abutments, the individualized CAD/CAM-manufactured abutments as well as the novel prefabricated bonding base abutments showed high mechanical stiffness and strength under quasistatic load. Especially, the monolithic design of the RNC-crowns in combination with bonding to titanium
8
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
abutments seems to be a promising prosthetic unit for fixed implant-supported reconstructions. ACKNOWLEDGMENT The authors would like to acknowledge Institute Straumann AG, Basel, Switzerland, for supplying the dental implants, prosthetic abutment components, and CAD/CAM-restoration blanks for RNC crowns used in this investigation. References 1. Beuer F, Stimmelmayr M, Gueth JF, Edelhoff D, Naumann M. In vitro performance of full-contour zirconia single crowns. Dent Mater 2012; 28:449–456. 2. Koller M, Arnetzl GV, Holly L, Arnetzl G. Lava ultimate resin nano ceramic for CAD/ CAM: customization case study. Int J Comput Dent 2012; 15:159–164. 3. Rekow ED, Silva NR, Coelho PG, Zhang Y, Guess P, Thompson VP. Performance of dental ceramics: challenges for improvements. J Dent Res 2011; 90:937–952. 4. Schoenbaum TR. Dentistry in the digital age: an update. Dent Today 2012; 31:108–113. 5. Patel N. Integrating three-dimensional digital technologies for comprehensive implant dentistry. J Am Dent Assoc 2010; 141(Suppl 2):20S–24S. 6. Roberts M. Strategies for integrating new restorative materials with digital technology and sound restorative principles. Compend Contin Educ Dent 2013; 34:52–57, 59. 7. Griffin JD Jr. Combining monolithic zirconia crowns, digital impressioning, and regenerative cement for a predictable restorative alternative to PFM. Compend Contin Educ Dent 2013; 34:212–222. 8. Christensen RP, Ploeger BJ. A clinical comparison of zirconia, metal and alumina fixed-prosthesis frameworks veneered with layered or pressed ceramic: a three-year report. J Am Dent Assoc 2010; 141:1317–1329.
9. Heintze SD, Rousson V. Survival of zirconia- and metalsupported fixed dental prostheses: a systematic review. Int J Prosthodont 2010; 23:493–502. 10. Beuer F, Schweiger J, Eichberger M, Kappert HF, Gernet W, Edelhoff D. High-strength CAD/CAM-fabricated veneering material sintered to zirconia copings – a new fabrication mode for all-ceramic restorations. Dent Mater 2009; 25:121– 128. 11. Guess PC, Bonfante EA, Silva NR, Coelho PG, Thompson VP. Effect of core design and veneering technique on damage and reliability of Y-TZP-supported crowns. Dent Mater 2013; 29:307–316. 12. Gigandet M, Bigolin G, Faoro F, Burgin W, Bragger U. Implants with original and non-original abutment connections. Clin Implant Dent Relat Res 2012. Doi: 10.1111/ j.1708-8208.2012.00479.x 13. Ferrario VF, Sforza C, Serrao G, Dellavia C, Tartaglia GM. Single tooth bite forces in healthy young adults. J Oral Rehabil 2004; 31:18–22. 14. Ferrario VF, Sforza C, Zanotti G, Tartaglia GM. Maximal bite forces in healthy young adults as predicted by surface electromyography. J Dent 2004; 32:451–457. 15. Kim JH, Lee SJ, Park JS, Ryu JJ. Fracture load of monolithic CAD/CAM lithium disilicate ceramic crowns and veneered zirconia crowns as a posterior implant restoration. Implant Dent 2013; 22:66–70. 16. Martinez-Rus F, Ferreiroa A, Ozcan M, Bartolome JF, Pradies G. Fracture resistance of crowns cemented on titanium and zirconia implant abutments: a comparison of monolithic versus manually veneered all-ceramic systems. Int J Oral Maxillofac Implants 2012; 27:1448–1455. 17. Leutert CR, Stawarczyk B, Truninger TC, Hammerle CH, Sailer I. Bending moments and types of failure of zirconia and titanium abutments with internal implant-abutment connections: a laboratory study. Int J Oral Maxillofac Implants 2012; 27:505–512. 18. Martin MP. Material and clinical considerations for fullcoverage indirect restorations. Compend Contin Educ Dent 2012; 33(Spec Issue 6):2–5. quiz 6.
ABSTRACT Background: Recent technical development allows the digital manufacturing of monolithic reconstructions with highperformance materials. For implant-supported crowns, the fixation requires an abutment design onto which the reconstruction can be bonded. Purpose: The aim of this laboratory investigation was to analyze stiffness, strength, and failure modes of implant-supported, computer-assisted design and computer-aided manufacturing (CAD/CAM)-generated resin nano ceramic (RNC) crowns bonded to three different titanium abutments. Materials and Methods: Eighteen monolithic RNC crowns were produced and loaded in a universal testing machine under quasi-static condition according to DIN ISO 14801. With regard to the type of titanium abutment, three groups were defined: (1) prefabricated cementable standard; (2) CAD/CAM-constructed individualized; and (3) novel prefabricated bonding base. Stiffness and strength were measured and analyzed statistically with Wilcoxon rank sum test. Sections of the specimens were examined microscopically. Results: Stiffness demonstrated high stability for all specimens loaded in the physiological loading range with means and standard deviations of 1,579 1 120 N/mm (group A), 1,733 1 89 N/mm (group B), and 1,704 1 162 N/mm (group C). Mean strength of the novel prefabricated bonding base (group C) was 17% lower than of the two other groups. Plastic deformations were detectable for all implant-abutment crown connections. Conclusions: Monolithic implant crowns made of RNC seem to represent a feasible and stable prosthetic construction under laboratory testing conditions with strength higher than the average occlusal force, independent of the different abutment designs used in this investigation. KEY WORDS: abutment connection, dental implants, DIN ISO 14801, failure mode, resin nano ceramic (RNC), stiffness, strength, titanium
materials.1–3 Recently, the entire prosthetic fabrication process of implant-supported reconstructions has been introduced in a complete digital workflow even without any physical models.4 In addition, the production of the suprastructure can be simplified by the option to connect crowns to prefabricated or individually customized abutments.5 In fixed implant prosthodontics, it is of interest to transfer the developments of digital computerassisted design and computer-aided manufacturing (CAD/CAM) technology to the fabrication of reconstructions with constant quality and a reasonable costbenefit ratio.6 Therefore, the demand arises for crowns made of high-strength monolithic materials, which are produced in a streamlined workflow and bonded to
INTRODUCTION Technical development in the field of digital dental medicine has opened the opportunity for the manufacturing of reconstructions using high-performance *Division of Fixed Prosthodontics, School of Dental Medicine, University of Bern, Bern, Switzerland; †Department of Prosthetic Dentistry, Center for Dental and Oral Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany; ‡Institute for Surgical Technology and Biomechanics, University of Bern, Bern, Switzerland Corresponding Author: Dr. Tim Joda, MSc, Division of Fixed Prosthodontics, School of Dental Medicine, University of Bern, Freiburgstr. 7, Bern CH-3010, Switzerland; e-mail: [email protected] © 2014 Wiley Periodicals, Inc. DOI 10.1111/cid.12215
1
2
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
TABLE 1 Physical Characteristics and Properties of Resin Nano Ceramic (RNC) Material
Density (g/cm3) Flexural strength (MPa0
RNC
2.1 204 1 19
Manufacturer information: Lava Ultimate Restorative, 3M ESPE, Neuss, Germany.
abutment substructures for one-piece screw-retained implant reconstructions.7 The potential of monolithic materials and their use especially in the functional zone of posterior regions has been under continuous investigation.3 Despite significant improvements in material properties, the clinical outcome of all-ceramic crowns still fails to match the success rates of metal ceramic-fixed reconstructions.8,9 In tooth-supported reconstructions, the fabrication method of CAD/CAM-constructed veneering material sintered to zirconia copings or full-contour zirconia crowns seem to fulfill high-loading capacities but also initiate excessive wear at natural antagonists.1,10,11 A recently defined material class – resin nano ceramic (RNC) – composed of bonded zirconia-silica nanoparticles, clustered and embedded in a cross-linked resin matrix, has been introduced for the CAD/CAM fabrication of fixed reconstructions (Lava Ultimate Restorative, 3M ESPE, Neuss, Germany).2 In preclinical tests, RNC exhibited a comparable fracture resistance to glass ceramics and a functional balance to enamel structures with a flexural modulus in the same range as dentin (Table 1). However, at present, no valid data are available in the scientific literature related to stiffness and strength of monolithic implant-supported reconstructions with RNC. The aim of this in vitro investigation was to compare stiffness, strength, and the failure modes of trans-occlusal, screw-retained, single-unit monolithic crowns made of RNC connected to two prefabricated and one individualized titanium abutment. The null hypothesis was that no difference exists in the mean stiffness, mean strength, and failure mode between the various abutment crown groups. MATERIALS AND METHODS Experimental Groups All specimens were designed as trans-occlusal, screwretained, one-piece reconstructions. The setup consisted
of a CAD/CAM-generated monolithic RNC-crown bonded to different types of abutments on soft tissue-level implants (Tissue Level Implant SLA RN 4.1 mm × 12 mm, Institute Straumann AG, Basel, Switzerland). In contrast to a naturally tooth-like appearance, the RNC crowns were fabricated as semi-sphere reconstructions with regard to the specific guidelines of DIN ISO 14801 (Dentistry-fatigue test for endosseous implants, International Organization for Standardization 2005, Geneva, Switzerland). Overall, the study included 18 test specimens, divided into 3 groups (Figure 1): •
•
•
Group A: n = 6 RNC crowns bonded to prefabricated standard titanium abutments (synOcta cementable, Institute Straumann AG); Group B: n = 6 RNC crowns bonded to individualized CAD/CAM-manufactured titanium abutments (CARES, Institute Straumann AG); Group C: n = 6 RNC crowns bonded to novel prefabricated bonding base titanium abutments (Variobase, Institute Straumann AG).
For manufacturing of the prosthetic RNC crowns, one master abutment per group was determined. The master abutments were mounted to corresponding implant analogues and digitalized with a laboratory CAD/ CAM scanner system (CARES CS2 Scanner, Institute Straumann AG). Three-dimensional virtual simulation and construction of the crowns were performed with a software program (CARES Visual Software 8.0, Institute Straumann AG). RNC blanks were used for the fabrication of the crowns in an external milling center (Straumann CAD/CAM GmbH, Markkleeberg, Germany). After milling, the RNC reconstructions were cleaned with 95% ethanol. Then, the titanium abutment types of groups A, B, and C were prepared with a specialized primer and bonded to the monolithic RNC crowns with resin cement (Panavia F 2.0, Kuraray Noritake Dental Inc., Tokyo, Japan) under laboratory conditions. Mechanical Resistance The preparation of the specimens for mechanical testing was performed according to a previously published article.12 Briefly summarized, the implants were perpendicularly embedded into a clamping device with
Implant-Supported Monolithic RNC Crowns
3
Figure 1 Study flowchart presenting a total of 18 specimens with 3 groups of different types of titanium abutments (Institute Straumann AG, Basel, Switzerland) bonded to monolithic RNC crowns (Lava Ultimate Restorative, 3M ESPE, Neuss, Germany). RNC = resin nano ceramic.
polymethyl methacrylate (PMMA) (EuroPlex SDX, Röhm AG, Brüttisellen, Switzerland) (Figure 2). The material is characterized by a modulus of elasticity higher than 3 GPa. The position was 3.0 mm 1 0.5 mm underneath the top of the implant neck in order to mimic physiologic bone loss according to DIN ISO 14801. The prepared monolithic one-piece reconstructions were mounted with 35 Ncm onto the dental implants using a manual torque control ratchet. Afterward, the specimens were clamped at a 30° 1 2° angle to the loading direction of the testing machine. Quasi-static load was applied with 1 mm/minute (Figure 3). Forces and displacements were recorded, and stiffness was defined as the maximum slope of a moving linear regression with a bandwidth that maximizes the correlation coefficient. Strength was documented as the maximal force until a decrease of 20% of the loadbearing capacity was noted.
(A)
(B)
Moreover, the mobility of implant prosthetic components was verified. In order to visualize the characteristics of failure modes and material alterations, crosssections were prepared for examination under polarized light. Specimens of each test group were embedded in methyl methacrylate and cut through the center in the direction of the loading zone into approximately 400-μm thick ground sections using a slow-speed diamond saw (Varicut VC-50, Leco, Munich, Germany). Polishing started with P100 grit silicon carbide paper and went through four levels, finishing with P600 grit. This procedure was followed by finer surface treatment with 0.05 micron alumina paste. Statistical Analysis Descriptive statistics as mean, standard deviation, and minimum and maximum of stiffness and strength were provided for the tested specimens of synOcta, CARES,
(C)
Figure 2 Technical preparation sequence of the test specimens with a specialized application tool (A), embedded in a clamping device with polymethyl methacrylate and placement 3.0 mm 1 0.5 mm underneath the top of the implant neck (B) as well as mounted RNC reconstruction (C). RNC = resin nano ceramic.
4
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Figure 3 Illustration of the testing set up according to the guidelines of DIN ISO 14801 (1 = implant reconstruction; 2 = PMMA embedding; 3 = embedding cup; 4 = fixation device; 5 = load cell; and 6 = piston). PMMA = polymethyl metharylate.
and Variobase abutments. Calculated stiffness and strength were statistically compared with Wilcoxon rank sum test. The level of statistical significance was set at p < .05. RESULTS None of the tested specimens displayed any obvious fractures after load application. All implant-abutment crown connections remained stable and no movement could be detected. Moreover, specimens of the three tested groups did not show any visible defects macroscopically. One specimen of the CARES abutment group fractured the RNC crown. Typical load displacement curves for the implant-abutment reconstructions with synOcta, CARES, and Variobase are shown in Figure 4.
Figure 4 Typical load displacement curves for the RNC implant-abutment reconstructions with synOcta abutment (group A), CARES abutment (group B), and Variobase abutment (group C). RNC = resin nano ceramic.
Figure 5 Boxplot diagram for stiffness of synOcta abutments (group A), CARES abutments (group B), and Variobase abutments (group C) with statistically significant differences between groups A and B (p < .05) (Wilcoxon rank test corrected for multiple testing by the method of Holm).
In addition to the individual results for all specimens, the mean and standard deviations of stiffness and strength of each group are reported in Table 2. The computed stiffness ranged from 1,362 N/mm to 1,876 N/ mm, while strength ranged from 1,088 N to 1,504 N. First, a Kruskal–Wallis test was performed: one each with the variable strength and the variable stiffness. The p values for strength and stiffness showed a significant difference in the overall median in both calculations. In an additional test, each two groups were compared in order to find exact differences. The Wilcoxon rank sum test compared the median of the two variables between each group. Statistical calculations revealed that mean stiffness of the Variobase group was equivalent to the means of the two other groups, but that the mean of the SynOcta group was 9% lower than the one of the CARES group (Figure 5). Mean strength of the Variobase group was 17% lower than those of the two other groups and that difference was statistically significant (Figure 6). The Variobase bonding abutment demonstrated the most homogenous results with the narrowest ranges of stiffness and strength (Figures 5 and 6). Microscopic examination of the cross-sections obtained from groups A, B, and C presented plastic deformations of the implants and abutments including screws for all tested specimens in the direction of the load application without any cracks (Figures 7 and 8).
Implant-Supported Monolithic RNC Crowns
Figure 6 Boxplot diagram of strength of synOcta abutments (group A), CARES abutments (group B), and Variobase abutments (group C) with statistically significant differences between groups A and C (p < .05) and groups B and C (p < .05) (Wilcoxon rank test corrected for multiple testing by the method of Holm).
DISCUSSION The findings of the present in vitro investigation demonstrated consistently high stiffness and strength under quasi-static loading for all three titanium abutment configurations in combination with the bonded monolithic RNC suprastructures. However, inner-group comparisons revealed 9% lower stiffness for the CARES group and 17% lower strength for prefabricated Variobase abutments compared with prefabricated synOcta and individualized CARES abutments. Because of the use of polymethyl metharylate (PMMA) as a reproducible substitute of bone for anchoring the implant, there is
no clinically relevant target value for absolute stiffness. Nevertheless, stiffness remains an important comparative variable to evaluate the proper function of load-bearing structures such as implant-supported reconstructions. Physiologic chewing forces for a single tooth in the molar area have been reported to range from 200 N to 400 N.13,14 In this physiologic load range, the mean stiffness of the Variobase design was equivalent to those of the two other groups and confers therefore similar stability. Moreover, all specimens exhibited quasi-static strength well beyond this physiological loading range. The results of laboratory studies must be translated carefully to clinical conditions, with varying loading modes as well as large number of loading cycles. For instance, in contrast to physiological chewing forces, the selected loading protocol was quasi-static. However, the failure mode was plastic deformation of metallic parts, which is essentially strain rate independent. Accordingly, based on the presented results and the current clinical experience, it can be assumed that all evaluated specimens could withstand masticatory forces. The abutment design selected in this trial seems to have no essential clinical relevance, even though some differences were statistically significant between the groups. Sample size was calculated with n = 6 specimens per test group. The technical preparation sequence of the specimens was standardized with a specialized application tool. This procedure guaranteed exact and reproducible results for embedding and placing in the clamping device in accordance to the guidelines of DIN ISO 14801.
TABLE 2 Results Displaying Maximal Forces (N) and Stiffness (N/mm) for Each Specimen of Groups A, B, and C
#
1 2 3 4 5 6 Average SD
Group A
Group B
Group C
synOcta Ti Abutment
CARES Ti Abutment
Variobase Ti Abutment
max Force (N)
Stiffness (N/mm)
max Force (N)
Stiffness (N/mm)
max Force (N)
Stiffness (N/mm)
1,409 1,504 1,435 1,412 1,146 1,196 1,350 144
1,579 1,571 1,723 1,640 1,362 1,601 1,579 120
1,398 1,502 1,287 1,275 1,335 1,306 1,350 86
1,612 1,673 1,721 1,771 1,873 1,745 1,733 89
1,153 1,117 1,147 1,113 1,088 1,132 1,125 24
1,738 1,395 1,876 1,773 1,724 1,719 1,704 162
SD = standard deviation.
5
6
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Figure 7 Cross-sectional analysis for groups A, B, and C with plastic deformations of the implant, abutment including screw, and bonded RNC crown (not visible in the embedded acrylic material) in the direction of load application without any signs of fractures (photomicrographic examinations with original ×12.5 magnification). RNC = resin nano ceramic.
For comparison, no studies were found reporting on monolithic implant reconstructions made of RNC and their mechanical behavior in the dental literature. Two recent laboratory studies, however, investigated the fracture resistance of monolithic lithium disilicate implant crowns.15,16 At molar sites, Kim and colleagues compared strength after load application of full contour lithium
Figure 8 Cross-section group C: implant, Variobase titanium abutment including screw and bonded RNC crown with close-up views of pressure (A) and tension zones (B). RNC = resin nano ceramic.
disilicate crowns to veneered zirconia substructures that were either hand layered or heat pressed.15 All crowns and zirconia copings were CAD/CAM generated with a labside milling system (Cerec, Sirona Dental System GmbH, Bensheim, Germany) and luted to the abutment substructures using resin cement. The monolithic lithium disilicate implant-supported reconstructions demonstrated a fracture load of 3,852.1 N. Heterogeneous results were reported on the zirconia-based reconstructions with significantly higher fracture loads for heat-pressed crowns (5,229.3 N) in contrast to handlayered crowns (3,100.3 N). Because of the different study design, the results are not directly comparable with the finding of the present investigations.15 In the second in vitro investigation, CAD/CAMfabricated and pressed monolithic lithium disilicate crowns were connected to customized implant abutments made of titanium and zirconia with a design for maxillary incisor replacement. In general, mean fracture resistance was significantly higher for reconstructions with titanium than zirconia abutments (p > .0001). Implant-supported crowns fabricated with CAD/ CAM-constructed monolithic lithium disilicate crowns in combination with titanium abutments exhibited the highest fracture resistance (558.5 N). All other
Implant-Supported Monolithic RNC Crowns
abutment-crown combinations ranged between 340.3 N and 495.9 N.16 The reported results for fracture resistances in case of full-contour lithium disilicate implant crowns connected to titanium abutments were obviously lower compared with the findings of the present trial with monolithic RNC crowns, regardless of the used abutment type. The interpretation of these findings may lead to the speculation that the crowns design or the restoration material itself had a significant influence of the fracture resistance of the entire reconstruction. In addition, individualized titanium and zirconia abutments with and without prosthetic suprastructures were tested under static loading in an in vitro approach identical to this investigation.17 The results of the individualized titanium abutment group demonstrated enhanced stability of unrestored test specimens (1,249 N) compared with implant-abutment reconstructions with glass-ceramic crowns (762 N). These findings were contradictory to the results of the present trial. The measured strength for the group with individualized titanium abutments connected to glass-ceramic crowns was at least 363 N lower than the average strength for the titanium abutments bonded with RNC crowns tested in this study. Thus, these results clearly illustrate the powerful connection of the monolithic RNC reconstructions with both the prefabricated and the individualized abutment types. In the aforementioned study, a remarkably high number of fractures of the glass-ceramic crowns (60% of the titanium abutment-supported crowns) were observed; and moreover, 30% of the specimens showed a fracture of the abutment screw or even of the implant. Twenty percent of the specimens showed mobility of at least one prosthetic component. These results suggest that the weakest link is the crown. In contrast to these findings, the present investigation revealed no detectable fractures after quasi-static load application. Neither any implant prosthetic component nor the implant were cracked. Based on the elastic properties of metallic materials, plastic deformation of the implant/abutment complex was observed in all test specimens. Surprisingly, none of the tested titanium abutments or the bonding connection to the monolithic RNC crown was mobile after loading. The abutment screws were able to stabilize the implant/ abutment complex even after irreversible deformation of the implant shoulder. The implant/abutment complex, the screw connections as well as the bonding
7
techniques for the monolithic RNC crowns represented a strong mechanical unit. A possible explanation for the higher strength could be the specific physical properties of RNC, particularly in contrast to lithium disilicate or glass ceramic. The composition of bonded zirconia-silica nanoparticles in a cross-linked resin matrix had the potential to imitate a comparable flexural modulus in the same range as human dentin (Table 1). The implementation of new technologies and treatment concepts within the digital workflow enable the production of monolithic reconstructions in a CAD/CAM-based process with high-performance dental materials, such as RNC.18 Demanding laboratory work steps may therefore be streamlined and the materialspecific advantages are ensured because of standardized production quality by using industrially manufactured material blanks.4,5 For these reasons, fully anatomical reconstructions made of RNC open a broader application with an enlarged range of indications in fixed implant prosthodontics. Because of the in vitro design of the study, the present investigation had some limitations, as an artificially crown design and no consideration of thermal effects, that complicate the direct comparison and translation to clinical situations. Nevertheless, the results of quasi-static load testing provide valid information and reveal important tendencies for treatment concepts with this newly defined class of monolithic restoration material. However, subsequent trials including fatigue cycling and thermal testing should be initiated to investigate the performance and durability of implant-supported reconstructions made out of RNC in combination with the bonding agent. Moreover, further clinical studies with close follow-up observations are necessary to confirm these laboratory results. CONCLUSIONS Monolithic implant crowns made of RNC are stable prosthetic reconstructions under laboratory testing conditions with high quasi-static mechanical properties compared with physiological occlusal force. The prefabricated cementable abutments, the individualized CAD/CAM-manufactured abutments as well as the novel prefabricated bonding base abutments showed high mechanical stiffness and strength under quasistatic load. Especially, the monolithic design of the RNC-crowns in combination with bonding to titanium
8
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
abutments seems to be a promising prosthetic unit for fixed implant-supported reconstructions. ACKNOWLEDGMENT The authors would like to acknowledge Institute Straumann AG, Basel, Switzerland, for supplying the dental implants, prosthetic abutment components, and CAD/CAM-restoration blanks for RNC crowns used in this investigation. References 1. Beuer F, Stimmelmayr M, Gueth JF, Edelhoff D, Naumann M. In vitro performance of full-contour zirconia single crowns. Dent Mater 2012; 28:449–456. 2. Koller M, Arnetzl GV, Holly L, Arnetzl G. Lava ultimate resin nano ceramic for CAD/ CAM: customization case study. Int J Comput Dent 2012; 15:159–164. 3. Rekow ED, Silva NR, Coelho PG, Zhang Y, Guess P, Thompson VP. Performance of dental ceramics: challenges for improvements. J Dent Res 2011; 90:937–952. 4. Schoenbaum TR. Dentistry in the digital age: an update. Dent Today 2012; 31:108–113. 5. Patel N. Integrating three-dimensional digital technologies for comprehensive implant dentistry. J Am Dent Assoc 2010; 141(Suppl 2):20S–24S. 6. Roberts M. Strategies for integrating new restorative materials with digital technology and sound restorative principles. Compend Contin Educ Dent 2013; 34:52–57, 59. 7. Griffin JD Jr. Combining monolithic zirconia crowns, digital impressioning, and regenerative cement for a predictable restorative alternative to PFM. Compend Contin Educ Dent 2013; 34:212–222. 8. Christensen RP, Ploeger BJ. A clinical comparison of zirconia, metal and alumina fixed-prosthesis frameworks veneered with layered or pressed ceramic: a three-year report. J Am Dent Assoc 2010; 141:1317–1329.
9. Heintze SD, Rousson V. Survival of zirconia- and metalsupported fixed dental prostheses: a systematic review. Int J Prosthodont 2010; 23:493–502. 10. Beuer F, Schweiger J, Eichberger M, Kappert HF, Gernet W, Edelhoff D. High-strength CAD/CAM-fabricated veneering material sintered to zirconia copings – a new fabrication mode for all-ceramic restorations. Dent Mater 2009; 25:121– 128. 11. Guess PC, Bonfante EA, Silva NR, Coelho PG, Thompson VP. Effect of core design and veneering technique on damage and reliability of Y-TZP-supported crowns. Dent Mater 2013; 29:307–316. 12. Gigandet M, Bigolin G, Faoro F, Burgin W, Bragger U. Implants with original and non-original abutment connections. Clin Implant Dent Relat Res 2012. Doi: 10.1111/ j.1708-8208.2012.00479.x 13. Ferrario VF, Sforza C, Serrao G, Dellavia C, Tartaglia GM. Single tooth bite forces in healthy young adults. J Oral Rehabil 2004; 31:18–22. 14. Ferrario VF, Sforza C, Zanotti G, Tartaglia GM. Maximal bite forces in healthy young adults as predicted by surface electromyography. J Dent 2004; 32:451–457. 15. Kim JH, Lee SJ, Park JS, Ryu JJ. Fracture load of monolithic CAD/CAM lithium disilicate ceramic crowns and veneered zirconia crowns as a posterior implant restoration. Implant Dent 2013; 22:66–70. 16. Martinez-Rus F, Ferreiroa A, Ozcan M, Bartolome JF, Pradies G. Fracture resistance of crowns cemented on titanium and zirconia implant abutments: a comparison of monolithic versus manually veneered all-ceramic systems. Int J Oral Maxillofac Implants 2012; 27:1448–1455. 17. Leutert CR, Stawarczyk B, Truninger TC, Hammerle CH, Sailer I. Bending moments and types of failure of zirconia and titanium abutments with internal implant-abutment connections: a laboratory study. Int J Oral Maxillofac Implants 2012; 27:505–512. 18. Martin MP. Material and clinical considerations for fullcoverage indirect restorations. Compend Contin Educ Dent 2012; 33(Spec Issue 6):2–5. quiz 6.
