Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema

Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns G.T. Rocca a,∗ , P. Sedlakova a , C.M. Saratti a , R. Sedlacek b , L. Gregor a , N. Rizcalla a , A.J. Feilzer c , I. Krejci a a

Division of Cariology and Endodontology, School of Dentistry, University of Geneva, Geneva, Switzerland Laboratory of Biomechanics, Department of Mechanics, Biomechanics and Mechatronics, Faculty of Mechanical Engineering, Czech Technical University, Prague, Czech Republic c ACTA, Amsterdam, The Netherlands b

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objective. To evaluate the influence of different types of modifications with resin on fatigue

Received 13 January 2016

resistance and failure behavior of CAD–CAM resin nano ceramic (RNC) restorations for max-

Received in revised form

illary first premolars.

19 July 2016

Methods. Sixty standardized resin composite root dies received CAD–CAM RNC endocrowns

Accepted 3 September 2016

(n = 30) and crowns (n = 30) (Lava Ultimate, 3M Espe). Restorations were divided into six

Available online xxx

groups: full anatomic endocrowns (group A) and crowns (group D), buccal resin veneered

Keywords:

and crowns (group F) with a central groove resin filling. A nano-hybrid resin composite was

CAD/CAM

used to veneer the restorations (Filtek Supreme, 3M Espe). All specimens were first submit-

endocrowns (group B) and crowns (group E) and buccal resin veneered endocrowns (group C)

Endocrown

ted to thermo-mechanical cyclic loading (1.7 Hz, 49 N, 600 000 cycles, 1500 thermo-cycles)

Fractography

and then submitted to cyclic isometric stepwise loading (5 Hz) until completion of 105 000

Veenering

cycles or failure after 5000 cycles at 200 N, followed by 20 000 cycles at 400 N, 600 N, 800 N,

Endodontically treated teeth

1000 N and 1200 N. In case of fracture, fragments were analyzed using SEM and modes of

Fatigue

failure were determined. Results were statistically analyzed by Kaplan–Meier life survival

Lava Ultimate

analysis and log rank test (p = 0.05). Results. The differences in survival between groups were not statistically significant, except between groups D and F (p = 0.039). Endocrowns fractured predominantly with a mesio-distal wedge-opening fracture (82%). Partial cusp fractures were observed above all in crowns (70%). Analysis of the fractured specimens revealed that the origin of the fracture was mainly at the occlusal contact points of the stepwise loading. Significance. Veneering of CAD–CAM RNC restorations has no influence on their fatigue resistance except when monolithic crowns are modified on their occlusal central groove. © 2016 The Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

∗ Corresponding author at: Department of Cariology and Endodontics, School of Dentistry, 19 Rue, Barthélémy Menn, 1205 Geneva, Switzerland. Fax: +41 22 3794102. E-mail address: [email protected] (G.T. Rocca). http://dx.doi.org/10.1016/j.dental.2016.09.024 0109-5641/© 2016 The Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS

xxx.e2

d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

1.

Introduction

CAD–CAM (computer-aided design/computer-aided machining) technology for dental purposes has highly evolved in the last few years. Hardware has become cheaper, software has become user friendly, fabrication is now faster and the milled workpieces are more accurate in respect of the anatomic form and dimensions as well as of the fit of the margins [1–3]. Early in-vivo performances of CAD–CAM restorations are also encouraging [4–6]. As a consequence, a wide range of new tooth-colored ceramic and resin based blocks are now available on the market. Ceramic materials, especially lithium-disilicate reinforced ones, display better mechanical (flexural strength, hardness, thoughness, wear resistance) and optical (translucency, opalescence, gloss) properties than the particulate filled resin materials. Polymer based materials are more interesting with regard to the ease of fabrication and practical clinical features such as the possibility of surface modification for functional or esthetic reasons by safe intraoral repair, the latter accomplished by surface sandblasting, while toxic hydrofluoric acid is normally needed for ceramic repair [7]. Some authors have also reported an optimal fatigue resistance of restorations made out of resin composite blocks, claiming their stress adsorbing properties [8–10]. Recently, resin materials with high ceramic fillers content have been introduced in the attempt to obtain intermediate properties between classical particulate-filled resins and ceramics. Lava Ultimate (3M Espe, St. Paul, MN, USA) is one of the firstborns of these “hybrid” materials. It has been introduced by the manufacturer as a resin nano-ceramic (RNC) material as the dimethacrylate resin matrix is charged with silica and zirconia nano-fillers to an extent of approximately 80% by weight. Improved flexural strength and toughness [8] compared to previous resin composite blocks (Paradigm MZ100, 3M Espe) and similar fatigue resistance [10] as well as fracture strength to glass-ceramics [11] have been reported for Lava Ultimate restorations. Some concerns arise about the esthetics of this material, as blocks are monochromatic and an esthetic modification of the raw workpiece after milling may be necessary depending on esthetic needs of the patient. This modification may be accomplished by replacing 1–2 mm of the buccal and/or occlusal part of the CAD–CAM restoration with an esthetic resin composite layer [7]. Restorative hybrid resins are often used for buccal veenering while stained resins with flowable consistency are used to mimic the occlusal fissure shade and anatomy. The adhesive link between the CAD–CAM core and the resin composite which is stratified on it has recently been argued. Lava resin blocks are industrially polymerized and display a high degree of monomer conversion resulting in a low amount of free radicals [12]. A mechanical pre-conditioning of the surface by high-pressure air abrasion or silica coating has proved to have a positive impact on micromechanical retention and adhesive strength [13–15]. Likewise, the use of a silane and a bonding resin as chemical intermediate agents over the conditioned CAD–CAM restoration surface seems today mandatory to reach an adequate bond strength with the veneering resin composite [13,15]. The objective of the modification is to improve the esthetic aspect of the CAD–CAM restoration without compromising its

mechanical properties. So far, the effect of such a structural modification on mechanical performances of CAD–CAM resin restorations has not yet been investigated. In fact, one of the major advantages of CAD–CAM workpieces in comparison to the classical lab-made ones is that they are milled from resin blocks which are fabricated under standardized and controlled high-pressure/high-temperature polymerization conditions. The resin composite produced is highly homogeneous and its mechanical properties are superior to the chair-side photopolymerized resin counterparts [12,16]. Replacing a part of the CAD–CAM resin restoration with a chair-side restorative resin composite could jeopardize the mechanical integrity of the milled restoration. Moreover, it is reasonable to suppose that this procedure could have a potential higher negative impact on crown restorations than for endocrowns as their thickness is limited by the presence of a core inside. The aim of this paper was to test in-vitro the fatigue behavior and the fracture mode of veneered CAD–CAM RNC crowns and endocrowns for upper premolars. Buccal and occlusal modifications of the restorations after cutting-back were tested. It was hypothesized that (a) buccal resin veneering of CAD–CAM RNC monolithic restorations has a negative impact on their mechanical performances and (b) this impact is higher if an occlusal composite resin filling of the central groove is introduced.

2.

Materials and methods

2.1.

Endocrowns/crowns fabrication

An extracted maxillary first premolar tooth was chosen as a master model. An optical impression of the crown (Bluecam, Cerec) was used to fabricate the crown anatomy of the CAD/CAM restorations using the software Cerec 4.0 in Biogeneric Copy Design mode. Then the premolar tooth model was cut at the CEJ and prepared for an endocrown restoration. An optical impression of the cavity was taken and the full anatomic CAD/CAM endocrown restorations were milled (n = 10, full anatomic endocrowns, control group A) (Fig. 1a). Endocrowns were 7.4-mm thick at the mesio-distal sulcus, including the 2.8 mm of the endo-core. Then, new endocrown restorations were fabricated with the same procedure but digitally cut-back with the Cerec software (6 × 5 mm, 1-mm deep) on the buccal side (n = 10, buccal veneered endocrowns, group B) and equally reduced on the buccal side but with a 1 × 3 mm, 1-mm deep occlusal groove (n = 10, buccal veneered endocrowns with an occlusal central groove, group C). Thereafter, a 2.8-mm high resin composite core was built up on the premolar root model and prepared for a crown restoration, without touching the outer limits (chamfer) of the former preparation for the endocrown. An optical impression of this core was taken (Fig. 1b) and the full anatomic CAD/CAM crown restorations were milled with the Cerec machine basing on the same Biogeneric Copy of group A (n = 10, full anatomic crowns, control group D). Crowns had standard dimensions and a thickness of 1.8 mm at the mesio-distal sulcus. New CAD/CAM crown restorations were then fabricated with the same procedure but with 6 × 5 mm, 1-mm deep buccal reduction (n = 10, buccal veneered crowns, group E) and with equal

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

2.3.

xxx.e3

Luting of the endocrowns/crowns

The intaglio surface of the CAD/CAM restorations as well as the root dies cavities were sandblasted with 27 ␮m aluminumoxide powder at about 0.2 MPa pressure for 5 s (MicroEtcher CD-Intraoral Sandblaster, Danville Materials, San Ramon, CA 94583, USA), rinsed and dried. The conditioned surfaces were then treated with an adhesive system (Scotchbond Universal Adhesive, 3M ESPE, Seefeld, Germany) without curing. A dualcuring resin composite (RelyX Ultimate, A2 shade, 3M ESPE, Seefeld, Germany) was used as luting cement. The restoration was seated in place first with manual pressure and then with the assistance of a specific ultra-sonic device (Cementation tip, EMS, Nyon, Switzerland). After removal of the excess cement, each restoration surface was light-cured for 60 s (buccal, occlusal and palatal) and margins were polished with discs of decreasing grain size, from coarse (80 ␮m) to superfine (20 ␮m) (SoFlex, 3M ESPE, Seefeld, Germany). Fig. 1 – (a, b) The Cerec scan of the premolar tooth model prepared for the endocrowns (a) and the crown (b) restorations. Symbols carved on the resin base are spatial references for the machine. They are necessary to mill the restoration from the Bio-copy scan of the premolar master model crown captured earlier. (c) Types of preparations for the resin modifications: full anatomical (no modifications) for groups A and D, buccal prepared (B and E) and occlusal–buccal prepared restorations (C and F). Measures are expressed in millimetres.

buccal reduction and above an occlusal 1 × 3 mm, 1-mm deep groove (n = 10, buccal veneered crowns with an occlusal central groove, group F) (Fig. 1c).

2.2.

Root dies fabrication

At the beginning, a trasparent silicon mold (Memosil 2, Heraeus Kulzer, Hanau, Germany) was created from the premolar tooth model and cut horizontally 1 mm over the CEJ. For each specimen, a nano-hybrid resin composite (Clearfil Majesty Posterior, Kuraray, Okayama, Japan) with an elastic modulus similar to that of human dentin (E-modulus 22 GPa) was inserted in the silicon mold to fabricate the root dies. Then, each restoration was isolated with glycerine, seated over the resin inside the silicone mold in order to stamp on this unpolymerized resin the shape of its inner-marginal area. The resin die simulating the root was then polymerized by using a LED lamp (Bluephase, Ivoclar-Vivadent, Schaan, Liechtenstein) through the transparent silicon mold. Root dies for endocrowns had 2.8-mm deep central pit corresponding to the endo-core of the restoration, surrounded by a chamfer preparation of around 1,2 mm. The core of the root dies for crown preparation was 2.8-mm high with the same chamfer preparation. All root dies were vertically fixed on a metallic holder. The base was embedded with self curing PMMA resin (Technovit 4071, Heraeus Kulzer, Hanau, Germany) till 1 mm under the artificial CEJ to complete the root stabilization.

2.4.

Veneering of the endocrowns/crowns

The grinded buccal and occlusal surfaces of specimens of groups B, C, E and F were submitted to airborne-particle abrasion with 27 ␮m aluminum oxide particles at about 0.2 MPa pressure for 5 s, rinsed and dried. The adhesive system (Scotchbond Universal Adhesive) was applied on all sandblasted surfaces and light-cured. A nanofilled resin composite (Filtek Supreme XTE Universal Restorative, A3B, 3M ESPE, Seefeld, Germany) was veneered on restorations’ buccal cavity in one increment and photo-polymerized for 30 s. It was then polished with discs of decreasing grain size. The occlusal grooves of groups C and F were filled with a flowable resin composite (Filtek Supreme XTE Flowable A3, 3M ESPE, Seefeld, Germany) and photo-polymerized for 30 s (Fig. 2 and Table 1).

2.5.

Thermo-mechanical fatigue loading

After 24 h the stress test was carried out with an established thermo-mechanical fatigue method, in a chewing simulator for Thermal Cycling and Mechanical Loading (TCML). All specimens were subjected to 600 000 cycles with 49 N axial occusal loading force applied with a ball of 3-mm diameter on the buccal cusp at a 1.7 Hz frequency following a one-half sine wave curve. By having the specimen holder mounted on a hard rubber disc, a sliding movement of the tooth is produced between the first contact on an inclined plane of the buccal cusp (mesial or distal) and the central fossa (Fig. 3a). A total of 1500 thermocycles (5 ◦ C to 50 ◦ C to 5 ◦ C) were performed simultaneously.

2.6.

Stepwise fatigue loading

After the TCML test all specimens were subjected to a cyclic loading test with a MTS Mini Bionix 858.02 servohydraulic testing system (Mini Bionix II, MTS, Eden Prairie, MN, USA) according to a stepwise loading method. The system was equipped with a load cell with a range of 0–2500 N. The chewing cycle was simulated by an isometric contraction. The loading member was a stainless steel ball of 3-mm diameter. Fatigue testing was carried out with unidirectional axial force. Because of the standardized anatomy, all restorations were

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

Scotchbond Universal Adhesive (3M ESPE)

Filtek Supreme XTE Universal (3M ESPE)

Filtek Supreme XTE Flowable (3M ESPE)

Clearfil Majesty Posterior (Kuraray)

√ Fracture toughness KIC (MPa m)

12,77a

2,02a

7,7a

1,95b

Not furnished

Not furnished

11,3a

1,84a

6,8a

1,67a

22a

1,06c

Bis-GMA, bis-phenol A diglycidylmethacrylate; TEGMA, triethyleneglycol dimethacrylate; HEMA, hydroxyethyl methacrylate; UDMA, urethane dimethacrylate; MDP, phosphate monomer dimethacrylate. a From manufacturers. b Bacchi et al. [52]. c Thomaidis et al. [53].

ARTICLE IN PRESS

RelyX Ultimate (3M ESPE)

Highly cross-linked resin matrix reinforced by 80 wt% of silane treated nano zirconia–silica particles agglomerated to clusters (0,6–10 ␮m) and individual silane bonded nano silica or zirconia particles (<20 nm) Adhesive resin cement consisting of methacrylate monomers, radiopaque, silanated fillers, initiator components, stabilizers, rheological additives, fluorescence dye dark cure activator for Scotchbond Universal adhesive MDP, HEMA VitrebondTM Copolymer Filler, ethanol, water, initiators, silane Bis-GMA, UDMA, TEGDMA, and bis-EMA resins, non-aggregated 20 nm silica filler, non-aggregated 4–11 nm zirconia filler, and aggregated zirconia/silica cluster filler (comprised of 20 nm silica and 4–11 nm zirconia particles) Bis-GMA, TEGDMA and Procrylat resins. Fillers of ytterbium trifluoride (particles 0.1–5.0 ␮m), a non-aggregated surface-modified 20 nm and 75 nm silica filler, a surface-modified aggregated zirconia/silica cluster filler (comprised of 20 nm silica and 4–11 nm zirconia particles) clusters (0.6–10 ␮m) Light-cure, nano-superfilled, radiopaque restorative posterior composite resin composed of nano and micro inorganic filler, silanated glass ceramic filler (average: 1.5 ␮m), surface treated alumina micro filler (average: 20 nm) Bis-GMA, TEGDMA, dl-camphorquinone, accelerators, pigments

E-modulus (GPa)

DENTAL-2840; No. of Pages 13

Lava Ultimate (3M ESPE)

Chemical compositiona

d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

Brand name (manufacturer)

xxx.e4

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

Table 1 – Materials used in the study.

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

xxx.e5

Fig. 2 – Schematic representation of the experimental groups. Endocrowns were 7.4-mm thick at the mesio-distal sulcus, including the 2.8 mm of the endo-core and the 1-mm deep occlusal filling in group C. Crowns had a thickness of 1.8 mm at the mesio-distal sulcus, including the 1-mm deep occlusal filling in group F. The E-modulus of restorative materials are: Clearfil Majesty Posterior (22 GPa), Lava Ultimate (12.77 GPa), Filtek Supreme XTE Universal (11.3 GPa), Filtek Supreme XTE Flowable (6.8 GPa).

adjusted in the same position with the loading sphere contacting both buccal and palatal cusps, halfway of the slope (Fig. 3b). The load varied sinusoidally between a nominal peak value F and 10% of this value (R = 0.1). The loading frequency was 5 Hz. The first 5000 cycles was a warm-upload at 200 N, followed by stages at 400, 600, 800, 1000 and 1200 N of a maximum of

20 000 cycles each. Specimens were loaded until fracture or to a maximum of 105 000 cycles and the number of endured cycles was registered. The integrity of the specimens was monitored throughout the test with a peak detector (Peak/Valley detector, MTS) which recognizes the difference between current loading and prescribed loading curve. The deviation was usually con-

Fig. 3 – A schematic representation of (a) the sliding movement imposed to specimens’ buccal cusp in the chewing simulator during the thermo-mechanical cyclic loading (TCML) and (b) the isometric stepwise cyclic loading test. (c) The different loading contacts are schematically represented in this occlusal view. The positioning of the contacts might slightly vary among the specimens. Typically, the sliding loading contact of the TCML (green) could be positioned on the mesial or distal inclined plane of the buccal cusp. Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS

xxx.e6

d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

Fig. 4 – Kaplan Mayer plotted survival curves of the experimental groups.

nected with excessive wear, accidental movements and first micro fractures inside the restoration.

2.7.

Fractography

After fracture all the specimens were visually examined in order to establish which fragments were suitable for fractographic analysis. The first examination of the broken specimens was performed using a stereomicroscope (SZX9, Olympus optical Co., Ltd., Tokyo, Japan). Characteristic features like compression curl, hackle and arrest lines were identified. Different magnifications (ranging from 6.3× to 50×) were used depending on the size of the characteristic marks detected. Angled illumination was used to better view the fracture surface. All recognizable features were photographed and documented. Scanning Electron Microscopy (SEM) (Digital SEM XL20, Philips, Amsterdam, Netherlands) was then used for a more detailed analysis of the fractured surfaces. In order to remove all of the impurities, all fragments were cleaned in an ultrasonic 10% sodium hypochlorite bath for 3 min, rinsed with water, dried and then fixed on the support for the microscope. The specimens were gold coated prior to the analysis with the SEM. Magnifications up to 2000× were used to obtain higher definition of identified crack features in selected areas of interest. The overall direction of crack propagation and failure origin(s) were systematically mapped for all specimens. The modes of fracture were analyzed by optical stereo microscopy and classified as (1) cuspal fracture or (2) split vertical fracture. Classification was based on an agreement between three examiners.

3.

Statistical analysis

Kaplan–Meier survival analysis was adopted for the visualization of time (cycles) to event in experimental groups; statistical

significance of differences in time to event between pairs of experimental groups was tested by means of log rank test. A p = 0.05 was chosen as a level of statistical significance in all analyses. Statistical analysis was computed using SPSS 22.0.0.1 (IBM Corporation, 2014).

4.

Results

None of the specimens revealed failures after the TCML test and no damage was detected at the stereomicroscope. After the stepwise fatigue test, restorations which survived the 105 000 cycles were 30% in group D, 20% in groups C, 10% in groups A, B and F and no one in group E. Differences in survival between the groups were not statistically significant except between groups D and F (p = 0.039) (Fig. 4 and Table 2). All restorations experienced non-reparable fractures. Though, different fracture paths were observed: specimens of groups A–C fractured predominantly with a mesio-distal vertical fracture which split the restoration (82%). Partial fractures propagating through restorations’ buccal or palatal cusps were also observed, above all in crowns of groups D–F (70%) (Table 3). The buccal resin veneer of groups B, C, E, and F was never involved in any of the crack paths. On the other hand, the occlusal resin filling of groups C and F was always debonded in fractured specimens. Fractographic analysis revealed two types of fatigue damages at the restorations surface: a sliding-contact worn surface over the buccal cusp due to the TCML and two impact induced rounded surfaces associated to the stepwise loading ball. In all fractured restorations the main origin of the fracture was located at the occlusal surface from the major contact loading area underneath the loading ball of the stepwise fatigue test, and propagated corono-apically (Figs. 5 and 6). Secondary minor events were noticed at the occlusal surface in some

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

xxx.e7

Fig. 5 – Group A, a representative split fracture of an endocrown restoration. (a) Occlusal view of the gold coated fractured specimen after repositioning of two main halves. The white circle indicates the worn surface provoked by the sliding movement during the TCML. The occlusal contacts of the stepwise fatigue load are labelled with two white stars, 1 and 2. Origin of the main crack varied depending on which one of the two contacts received the highest amount of load (in this case the buccal, 1). In this gold treated version of the specimen, characteristic marks left by the CAD–CAM machining burs are clearly evident all over the unpolished surface. (b) A schematic representation of the split fracture. The presence of a third minor fragment between two main halves gave to the fracture path a typical “Y” shaped form on a sagittal plane. (c) Picture from the stereomicroscope of the fractured surface. Borders of the third additional chip are evident (fragmented white line) (d) SEM large view of the fractured surface. (e) Details of the primary crack origin (white tip) in higher magnification. Hackle lines are clearly visible and they indicate the direction of crack propagation (white arrows). The big white arrow indicates an arrest line where the crack changes its direction. (f) Close up of the restoration surface. Two kinds of fatigue damages are visible, a smooth worn surface due to the sliding contact of the TCML (white circle) and a ring rough surfaces (star 1) generated by the impact of stepwise loading ball. (g) Higher magnification of an area of the fractured surface far from the main origin where the roughened restoration surface generated by the CAD–CAM burs (65 ␮m of granulometry) is clearly visible on top of the restoration. (h) Detail of the microscopic milling surface damages on the edge of the fractured surface.

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS

xxx.e8

d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

Table 2 – Pairwise post hoc comparisons with log rank test and medians for survival time. Log rank test A B C D E F

A

B

C

– p = 0.565 p = 0.753 p = 0.223 p = 0.843 p = 0.257

p = 0.565 – p = 0.896 p = 0.091 p = 0.523 p = 0.437

p = 0.753 p = 0.896 – p = 0.177 p = 0.713 p = 0.498

D p = 0.223 p = 0.091 p = 0.177 – p = 0.104 p = 0.022

E

F

p = 0.843 p = 0.523 p = 0.713 p = 0.104 – p = 0.249

p = 0.257 p = 0.437 p = 0.498 p = 0.022 p = 0.249 –

Medians for survival time (95% CI) A B C D E F

85 011 (62 046; 107 976) 65 029 (59 761; 70 297) 65 047 (45 461; 84 633) 86 091 (82 846; 89 336) 85 005 (59 740; 110 270) 52 908 (47 871; 57 945)

Table 3 – Types of macroscopic failure for the experimental groups. Palatal or buccal origin of the main crack under the occlusal load of the stepwise fatigue test varied depending on which one of the two contacts received the highest amount of load. A slight variation in specimen orientation could change the primary origin. A

B

C

D

E

Cusp fracture

2

1

1

4

7

7

Split fracture

7

8

7

3

3

2

specimens, originating from the multiple superficial defects of the CAD–CAM unpolished restoration surface.

5.

Discussion

In this in-vitro study the survival rate and the fracture mode of monolithic and modified CAD–CAM RNC crowns and endocrowns restorations were investigated. Results of the life table analysis showed that there were no significant differences among the tested groups with regard to fatigue resistance except between the monolithic crowns (group D) and the buccal–occlusal modified counterparts (group F). Thus, for crown restorations the first hypothesis was rejected and the second hypothesis was accepted. For endocrowns, both hypotheses were rejected. A maxillary first premolar was chosen as the tooth model firstly because the monochromatic restorations in the upper premolar region could require esthetic corrections. Also, the particular crown anatomy of an upper premolar which has pronounced mesio-distal furcations and an occlusal anatomy with a sharp angle between the cusps makes these teeth more susceptible to a mesio-distal split vertical fracture [17]. The abovementioned crown anatomy was transmitted to all CAD–CAM industrially processed RNC blocks by the Cerec 4.0 software in Biogeneric Copy Design mode. This procedure guaranteed extremely standardized crown and endocrown restorations in terms of dimensions and material properties among all specimens. Moreover, to further improve the standardization of the study set-up resin composite root dies were fabricated in series using a resin composite with a standardized silicon mold instead of prepared natural teeth. A resin composite with an elastic modulus (22 GPa) similar to that of the dentinal tissue was employed as substitute for dentin [9,11,18,19].

F

Specimens in this in-vitro set-up were fatigued in two separate phases. During the first phase, all of the restorations were subjected to a thermo-mechanical cycling loading in a computer-controlled masticator. In order to mimic invivo testing conditions, the low masticatory force of 49 N was applied for 600 000 cycles at 1.7 Hz, corresponding to approximately 2.5 years of mild clinical function (without peaks) in premolar region [20–22]. A more clinically relevant number of cycles of 106 at low masticatory loads – corresponding to approximately 5 years of function – has been suggested before [23]. Though, in present study, fatigue conditions were exacerbated later and, moreover, a recent clinical trial on composite resin posterior crowns reported different kinds of complications long before 5 years of clinical use [24]. Cyclic loading and water aging are known to degrade the fracture resistance of Lava Ultimate restorations [25,26] while thermocycling has a negative influence on the final strength of a resin composite [27]. During the second phase, fatigue testing conditions were “accelerated” by increasing stepwise the load as well as the frequency of the cycles. Similar conditions have been already applied in others in-vitro studies [9,28,29,30]. In the clinical setting, restorations are known to fail mainly under fatigue conditions [31,32]. At low masticatory loads – such as in the first phase of this study – process leading to failure may last millions of cycles. The rationale behind the second phase was to intensify the testing conditions in order to provoke restoration failure under fatigue conditions, over a certain number of cycles. Under these testing conditions the first degradation of the material may occur below critical failure stresses and involve the growth of subcritical defects at subcritical loads [26,33]. However, neither the load-to-fracture values nor the number of cycles-to-fracture obtained with this second phase of the fatigue test – the stepwise test – should be used to predict clinical mechanical limits of different restorations, due to the extreme and

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

xxx.e9

Fig. 6 – Group D, a representative cusp fracture of a crown restoration. (a) Occlusal view of the gold coated fractured specimen after repositioning of two main halves. The white circle indicates the worn surface provoked by the sliding movement during the TCML. The occlusal contacts of the stepwise fatigue load are labelled with two white stars (1 and 2). Origin of the main crack varied depending on which one of the two contacts received the highest amount of load (in this case the buccal, 1). A third additional fragment between two main halves originated close to the primary crack origin. (b) A schematic representation of the buccal cusp fracture on a sagittal plane. (c) A lateral view of the fractured specimen. (d) Picture from the stereomicroscope of the fractured surface. Borders of the third additional chip are evident (fragmented white line). (e) SEM large view of the fractured surface. The fracture path presents a distinct main crack front running from the top of the restoration to the restoration’s marginal areas. At the bottom of the fractured surface a compression curl is visible (white square). (f) The fractured surface in higher magnification. Big white tip indicates the origin of the main crack. A zone of permanent deformation is visible under the main loading contact. White arrows show the direction of the hackle lines underneath the origin. They indicate the direction of crack propagation, dcp. The big white arrow indicates an arrest line where the crack changes its direction, just before the resin core underneath the crown.

non-physiological testing conditions imposed on the materials. Analysis of the plotted Kaplan–Meyer survival curves shows that none of the groups can offer the longest survival rate (Fig. 4). Differences in survival between the groups were not statistically significant except between groups D and F (Table 2). Thus, when comparing the type of restoration, Lava Ultimate endocrowns and crowns were equivalent in terms of fatigue resistance in both the monolithic or the modified versions. These results are in agreement with Ramirez et al.

[34] and Magne et al. [10] who compared CAD–CAM resin endocrowns and crowns for upper incisors and lower molars respectively. Considering the impact of both kinds of resin modifications – buccal and occlusal – on monolithic CAD–CAM blocks, buccal resin veneering of the restorations with the nano-hybrid resin composite (Filtek Supreme XTE Universal Restorative, 3M ESPE) did not have an influence on life survival both for endocrowns and crowns. Specimens of modified groups B, C, E and F could accomplish the two phases of the fatigue

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS

xxx.e10

d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

test without any debonding of the resin veneer visible in the stereomicroscope. Also, fractographic analysis of the fractured specimens from these groups and the analysis of their mode of fracture excluded any involvement of this buccal modification in crack origin and growth. This result was not unexpected, as no direct loading or damage at the buccal veneer bonded interface occurred during any phase of the fatigue loading. Certainly the optimal bonding procedure between Lava Ultimate restorations and the restorative resin veneer [35] was able to counteract thermal and mechanical stresses during fatigue but the distance between this buccal resin veneer and the occlusal loads should have played a major role on veneer integrity, as stress distributed within a resin restoration is known to be concentrated around the loading contacts and not transmitted to peripheral areas [11,36,37]. A further demonstration is the fact that the occlusal resin filling of restorations of groups C and F was always debonded during fracture, as a consequence of the proximity with the occlusal origin of the overloading event. Thus, in this in-vitro set-up, replacing the buccal part of a CAD–CAM RNC restoration with a chair-side restorative resin composite does not jeopardize the mechanical integrity of the milled restoration. On the other hand, the occlusal resin filling (group F) weakened the monolithic crowns (group D). Clinically, this occlusal modification is accomplished to improve esthetics of monolithic CAD–CAM resin restorations, above all in posterior teeth of the lower arch. In clinics, the anatomy of the primary fissure of the raw restoration is underlined with a sharp bur, and a colored flowable resin is bonded inside this tiny cavity to simulate a naturally stained fissure. In this in-vitro study, dimensions of this occlusal central groove were accentuated (1 × 3 mm, 1-mm deep) versus the clinical reality for standardization purposes. As it was expected, this occlusal intervention had a weakening effect on fatigue resistance of crowns of group F due to a critical alteration of the restoration’s thickness just underneath the occlusal loads. Beforehand, Chen et al. [11] have already shown an increasing linear relation between thickness and fracture strength of Lava Ultimate restorations. Interestingly, that occlusal intervention was uninfluential for modified endocrowns of group C, if compared to unmodified thicker endocrowns of group A. A potential explication for that difference is that the occlusal thickness of crowns restorations of group F (around 1 mm without the 1-mm deep occlusal modification) is limited if compared to endocrowns of group C (around 6.5 mm without the modification) and, in proportion, this occlusal intervention was critical for fracture resistance only for thinner crown restorations of group F. Further in-vitro studies are needed to clarify this issue. In order to investigate the events which took place during the fracture, a fractographic analysis was carried out on all fragments. A combined stereo and scanning electron microscopy technique was used [38,39]. While stereomicroscope revealed a better 3D vision of the fractured surface, SEM images efficiently provided the information on characteristic fractographic markers such as hackle lines, arrest lines and wake hackle, indicators of the crack propagation direction. More than the half of fractured restorations (30 out of 52) experienced split vertical fractures which extended from the occlusal main fissure to approximately the mid-point of the mesial and distal marginal ridges (Fig. 5). The high incidence of

this mode of fracture was not unexpected due to the positioning of the loading sphere of the isometric stepwise test which was in contact with both buccal and palatal cusps, halfway of the occlusal slope. The main crack originated from the contact which received the higher amount of load. Nearby to the major impact and between the two main fragments a third chip formed regularly. The horizontal (mesio-distal) and vertical dimensions of this fragment may vary and it could extend deep inside the restoration. The presence of this fragment gave to the fracture path a typical “Y” shaped form on a sagittal section. The split mode of fracture was above all prevailing in endocrown restorations (82%) compared to crowns (30%) (Table 3). All these wedge-opening fractures progressed also in the root resin dies provoking drastic “root” breakdowns. This scenario should not be used to predict real clinical premolar root failures as crack propagation under fatigue mechanism (or fatigue crack growth) inside a material is known to be influenced by several inherent material properties – diameter and orientation of tubuli in dentin, filler population and size in resin composite [40,41] as an exemple – and, except for the elastic modulus, the resin composite used in this in-vitro setup to simulate the roots displays features totally different to dentin. Also, the presence in current specimens of a resin-toresin bonding interface instead of a dentin-to-resin one could have had an influence on crack propagation [40,42]. However, it is reasonable to suppose that the particular design of a monolithic endocrown restoration, which clinically extends deeply inside the root and beyond the CEJ with an endo-core, could expose the root to irreversible fracture if the crack has already split the restoration. The fractographic analysis of this kind of wedge-opening fractures revealed that the primary crack front originates at the occlusal surface of the restoration from the major contact loading area, and then propagates downwards through the Lava Ultimate endocrown, provoking a bulk fracture (Fig. 5). The main crack ran rapidly from top to bottom of the specimen changing its direction but without any interruption and with a relatively smooth crack front. Two kinds of fatigue damages were detected at the restorations surface, the worn surface over the buccal cusp due to the sliding contact of the TCML and two ring surfaces generated by the impact of stepwise loading ball (Figs. 5 and 6). In fractured specimens, surface and sub-surface damages due to stepwise loading were always visible in section as the primary crack origin lies constantly at the main occlusal load. In contrast, surface damage created by the sliding movement (TCML) was just visible as a smooth worn surface. About correlation with the fracture, there was no evidence of an association between this worn surface and the main origin of the crack. In many fractured specimens this surface was far from fracture’s origin and path. A zone of permanent deformation was always present underneath the main contact surface of the stepwise load, where the crack initiated. Except for the presence of this plastic deformation area, the fracture surface generated by the present compressive cyclic isometric load displayed analogous appearance and fractographic markers to fracture patterns of fractured Lava Ultimate endocrowns tested under higher single-cycle loads (more than 2000 N) in a recent study by Rocca et al. [43]. This analogy in the effects could indirectly mean that the dynamic stepwise loading test imposed in this study generates stresses on restoration surface

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

similar to static single-cycle loads and, thus, that this stepwise loading regimen is more load- than time-dependent. Moreover, in present study, restorations which break at the highest loads did not display rougher fractured surfaces and more crack branching than those which break at lower loads as it is described in previous research papers under different loading conditions [31,44]. About the extensive sub-surface damage made by the stepwise loading ball, it is important to note that small radii indenters such as the one used can generate intense point loads, which are more likely to create surface damages than bulk effects [45]. In the present study, dimensions of the indenter were limited by the peculiar occlusal anatomy of the upper premolars, which presents a small intercuspal angle. The rest of the fractured specimens displayed cuspal partial fractures which were always extensive and non-reparable (Fig. 6). This mode of fracture was above all prevailing in crown restorations (70%) compared to endocrowns (18%) (Table 3). In the majority of these partial fractures in crowns, the resin core was not or was minimally involved in the crack path. In fact, the crack front had its origin at the major loading occlusal contact. It passed through the first part of the CAD–CAM crown restoration vertically as the wedge opening fracture described before for endocrowns but, in front of the resin core, it could deviate buccally or palatally – depending if the origin of the crack was buccal or palatal respectively – and it ended its run close to the restoration margin. The presence of a stiffer core (22 GPa) underneath the RNC crown (12.77 GPa) could have contributed to this crack deflection. This effect has been already shown for multi-layered ceramic structures [46–48]. Clinically, if the crack avoids the dentin core and passes nearby without splitting the root that would mean the conservation of the tooth element. Fractographic analysis of these cuspal fractures (Fig. 6) revealed similar primary crack origin and fracture patterns of the wedge-opening fractures described before. In some fractured endocrowns and crowns, irrespective of the mode of fracture – cuspal or split fracture – multiple minor crack origins were also noticed at the occlusal surface, far from the occlusal main load (Fig. 5). These secondary events probably originated from the various superficial defects of the resin restoration rough surface and then propagated in a coronal-apical direction. These microscopic surface imperfections (machining defects) were generated by the CAD–CAM burs (around 65 ␮m of granulometry), during the milling phase of the resin restoration. The latter was not fine polished after milling to avoid any influence on specimens standardization. During the complex dynamic of the fast fracture, it is reasonable to suppose that, under the high tensile force generated by the loading ball positioned halfway of both cusps, these surface imperfections could have acted as weakening flaws producing several minor secondary crack fronts. It is important to note that those few restorations in which this phenomenon was noticed did not display different life survival after the fatigue loading. While authors were writing this manuscript, 3M ESPE removed the crown indication (and by that even endocrowns) for Lava Ultimate because of a higher-than-anticipated rate of debonding, whilst leaving the indication for inlays/onlays with an internal retentive design and veeners. In that announcement, no details were included about the proportion of

xxx.e11

this phenomenon, the type of concerned crowns (anterior/posterior crowns, restoration’s thickness) and, above all, about the adhesive strategy involved in debonded cases (type of cement, surfaces pre-treatment). Results of present study have shown an optimal behavior of this material in terms of fatigue resistance, without any final debonding. Shembish et al. in a recent in-vitro fatigue test [49] displayed similar outcomes for Lava Ultimate molar crowns bonded to resin composite standardized dies (Filtek Z100, 3M ESPE). The presence in our study and in the latter of a resin cavity as adhesive substrate to replace dentin has certainly improved the standardization of the set-up but it could have influenced also the long-terme adhesive performances of Lava Ultimate crowns and endocrowns. However, under analogous testing conditions (stepwise fatigue loading) Magne et al. [50] and Carvalho et al. [51] showed optimal fatigue resistance of Lava Ultimate crowns bonded to coronal dentin, without any restoration debonding but partial adhesive failures. At the best of authors’ knowledge, no clinical trials exist on Lava Ultimate crowns. Inasmuch as in previous in-vivo studies on survival of composite resin posterior crowns, such as the 3-years prospective study led by Jongsma et al. [24] debonding of restorations is normally encountered as a complication, it is reasonable to expect a similar outcome for a RNC material, which, from a material science perspective, still belongs to the resin composite category. More clinical research is needed to verify the long-term adhesive potential of this material in order to extend its indications to all non-retentive tooth cavities.

6.

Conclusion

Within the limits of this in-vitro study it was concluded that the modification of CAD–CAM RNC restorations for upper premolars with a restorative resin for esthetic purposes has no influence on their fatigue resistance except when monolithic crowns are modified at their occlusal surface. The prevailing mode of fracture for endocrown restorations was a split vertical failure while crowns fractured more with partial cusp fractures. This difference in failure behavior seems to be more related to the restoration design (presence of a core) than to the type of surface resin modification. Further in-vitro studies and clinical trials are needed to confirm these results.

Acknowledgments The authors wish to thank 3M Espe and Kuraray for their supply of the tested materials and Dr. Izabella Nerushay for the English proofreading.

references

[1] Akbar JH, Petrie CS, Walker MP, Williams K, Eick JD. Marginal adaptation of Cerec 3 CAD/CAM composite crowns using two different finish line preparation designs. J Prosthodont 2006;15:155–63. [2] Sato K, Matsumura H, Atsuta M. Relation between cavity design and marginal adaptation in a machine-milled ceramic restorative system. J Oral Rehabil 2002;29:24–7.

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS

xxx.e12

d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

[3] Reiss B. Occlusal surface design with CEREC 3D. Int J Comput Dent 2003;6:333–42. [4] Tamim H, Skierven H, Ekfeldt A, Rønold HJ. Clinical evaluation of CAD/CAM metal-ceramic posterior crowns fabricated from intraoral digital impressions. Int J Prosthodont 2014;27:331–7. [5] Syrek A, Reich G, Ranftl D, Klein C, Cerny B, Brodesser J. Clinical evaluation of all-ceramic crowns fabricated from intraoral digital impressions based on the principle of active wavefront sampling. J Dent 2010;38:553–9. [6] Otto T, Mörmann WH. Clinical performance of chairside CAD/CAM feldspathic ceramic posterior shoulder crowns and endocrowns up to 12 years. Int J Comput Dent 2015;18:147–61. [7] Rocca GT, Bonnafous FC, Rizcalla N, Krejci I. A technique to improve the esthetic aspects of CAD/CAM composite resin restorations. J Prosthet Dent 2010;104:273–5. [8] Ruse ND, Sadoun MJ. Resin-composite blocks for dental CAD/CAM applications. J Dent Res 2014;93:1232–4. [9] Magne P, Schlichting LH, Maia HP, Baratieri LN. In vitro fatigue resistance of CAD/CAM composite resin and ceramic posterior occlusal veneers. J Prosthet Dent 2010;104:149–57. [10] Magne P, Carvalho AO, Bruzi G, Anderson RE, Maia HP, Giannini M. Influence of no-ferrule and no-post buildup design on the fatigue resistance of endodontically treated molars restored with resin nanoceramic CAD/CAM crowns. Oper Dent 2014;39:595–602. [11] Chen C, Trindade FZ, de Jager N, Kleverlaan CJ, Feilzer AJ. The fracture resistance of a CAD/CAM Resin Nano Ceramic (RNC) and a CAD ceramic at different thicknesses. Dent Mater 2014;30:954–62. [12] Nguyen JF, Migonney V, Ruse ND, Sadoun M. Resin composite blocks via high-pressure high-temperature polymerization. Dent Mater 2012;28:529–34. [13] Stawarczyk B, Krawczuk A, Ilie N. Tensile bond strength of resin composite repair in vitro using different surface preparation conditionings to an aged CAD/CAM resin nanoceramic. Clin Oral Investig 2015;19:299–308. [14] Wiegand A, Stucki L, Hoffmann R, Attin T, Stawarczyk B. Repairability of CAD/CAD high-density PMMA- and composite-based polymers. Clin Oral Investig 2015;19:2007–13. [15] Ozcan M, Barbosa SH, Melo RM, Galhano GA, Bottino MA. Effect of surface conditioning methods on the microtensile bond strength of resin composite to composite after aging conditions. Dent Mater 2007;23:1276–82. [16] Pereira SG, Fulgêncio R, Nunes TG, Toledano M, Osorio R, Carvalho RM. Effect of curing protocol on the polymerization of dual-cured resin cements. Dent Mater 2010;26:710–8. [17] Sornsuwan T, Ellakwa A, Swain MV. Occlusal geometrical considerations in all-ceramic pre-molar crown failure testing. Dent Mater 2011;27:1127–34. [18] Zhao K, Wei YR, Pan Y, Zhang XP, Swain MV, Guess PC. Influence of veneer and cyclic loading on failure behaviour of lithium disilicate glass-ceramic molar crowns. Dent Mater 2014;30:164–71. [19] Wang Y, Katsube N, Seghi RR, Rokhlin SI. Statistical failure analysis of adhesive resin cement bonded dental ceramics. Eng Fract Mech 2007;74:1838–56. [20] Krejci I, Lutz F, Krejci D. The influence of different base materials on marginal adaptation and wear of conventional Class II composite resin restorations. Quintessence Int 1988;19:191–8. [21] Rocca GT, Gregor L, Sandoval MJ, Krejci I, Dietschi D. In vitro evaluation of marginal and internal adaptation after occlusal stressing of indirect class II composite restorations with different resinous bases and interface treatments.

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32] [33] [34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

“Post-fatigue adaptation of indirect composite restorations”. Clin Oral Investig 2012;16:1385–93. Kumagai H, Suzuki T, Hamada T, Sondang P, Fujitani M, Nikawa H. Occlusal force distribution on the dental arch during various levels of clenching. J Oral Rehabil 1999;12:932–5. Wiskott HW, Nicholls JI, Belser UC. Stress fatigue: basic principles and prosthodontic implications. Int J Prosthodont 1995;8:105–16. Jongsma LA, Kleverlaan CJ, Feilzer AJ. Clinical success and survival of indirect resin composite crowns: results of a 3-year prospective study. Dent Mater 2012;28:952–60. Al-Harbi FA, Ayad NM, ArRejaie AS, Bahgat HA, Baba NZ. Effect of aging regimens on resin nanoceramic chairside CAD/CAM material. J Prosthodont 2015;14. Belli R, Geinzer E, Muschweck A, Petschelt A, Lohbauer U. Mechanical fatigue degradation of ceramics versus resin composites for dental restorations. Dent Mater 2014;30:424–32. Ayatollahi MR, Yahya MY, Karimzadeh A, Nikkhooyifar M, Ayob A. Effects of temperature changes and beverage on mechanical and tribological properties of dental restorative composites. Mater Sci Eng C Mater Biol Appl 2015;54:69–75. Magne P, Knezevic A. Thickness of CAD-CAM composite resin overlays influences fatigue resistance of endodontically treated premolars. Dent Mater 2009;25:1264–8. Batalha-Silva S, de Andrada MA, Maia HP, Magne P. Fatigue resistance and crack propensity of large MOD composite resin restoration: direct versus CAD/CAM inlay. Dent Mater 2013;29:324–31. Kuijs RH, Fennis WM, Kreulen CM, Roeters FJ, Verdonschot N, Creugers NH. A comparison of fatigue resistance of three materials for cusp-replacing adhesive restorations. J Dent 2006;34:19–25. Lohabauer U, Belli R, Ferracane JL. Factors involved in mechanical fatigue degradation of dental resin composites. J Dent Res 2013;93:584–91. Ferracane JL. Resin-based composite performance: are there some things we can’t predict? Dent Mater 2013;29:51–8. Baran G, Boberick K, McCool J. Fatigue of restorative materials. Crit Rev Oral Biol Med 2001;12:350–60. Ramírez-Sebastià A, Bortolotto T, Cattani-Lorente M, Giner L, Roig M, Krejci I. Adhesive restoration of anterior endodoctically treated teeth: influence of post length on fracture strength. Clin Oral Investig 2014;18:545–54. Rinastiti M, Ozcan M, Siswomihardjo W, Busscher HJ. Immediate repair bond strengths of microhybrid, nanohybrid and nanofilled composites after different surface treatments. J Dent 2010;38:29–38. Nakamura T, Imanishi A, Kashima H, Ohyama T, Ishigaki S. Stress analysis of metal-free polymer crowns using the three-dimensional finite element method. Int J Prosthodont 2001;14:401–5. Ohlmann B, Gruber R, Eickemeyer G, Rammelsberg P. Optimizing preparation design for metal-free composite resin crowns. J Prosthet Dent 2008;100:211–9. Scherrer SS, Quinn JB, Quinn GD, Wiskott HW. Fractographic ceramic failure analysis using the replica technique. Dent Mater 2007;23:1397–404. Scherrer SS, Quinn GD, Quinn JB. Fractographic failure analysis of a Procera® AllCeram crown using stereo and scanning electron microscopy. Dent Mater 2008;25:1107–13. Soappman MJ, Nazari A, Porter JA, Arola D. A comparison of fatigue crack growth in resin composite dentin and the interface. Dent Mater 2007;23:608–14. Arola DD, Reprogel RK. Tubule orientation and the fatigue strength of human dentin. Biomaterials 2006;27:2131–40.

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024

DENTAL-2840; No. of Pages 13

ARTICLE IN PRESS d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e13

[42] Zhang Z, Beitzel D, Mutluay M, Tay FR, Pashley DH, Arola D. On the durability of resin-dentin bonds: identifying the weakest links. Dent Mater 2015;31:1109–18. [43] Rocca GT, Saratti CM, Cattani- Lorente M, Feilzer AJ, Scherrer SS, Krejci I. The effect of a fiber reinforced cavity configuration on load bearing capacity and failure mode of endodontically treated molars restored with CAD/CAM resin composite overlay restorations. J Dent 2015;43:1106–15. [44] Quinn JB, Quinn GD. Materials properties and fractography of an indirect dental resin composite. Dent Mater 2010;26:589–99. [45] Kelly JR, Rungruanganunt P, Hunter B, Vailati F. Development of a clinically validated bulk failure test for ceramic crowns. J Prosthet Dent 2010;104:228–38. [46] Lawn BR, Pajares A, Zhang Y, Deng Y, Polack MA, Llyod IK, et al. Materials design in the performance of all-ceramic crowns. Biomaterials 2004;25:2885–92. [47] Hermann I, Bhowmick S, Lawn BR. Role of core support material in veneer failure of brittle layer structures. J Biomed Mater Res B Appl Biomater 2007;82:115–21.

xxx.e13

[48] Scherrer SS, de Rijk WG. The fracture resistance of all-ceramic crowns on supporting structures with different elastic moduli. Int J Prosthodont 1993;6:462–7. [49] Shembish FA, Tong H, Kaizer M, Janal MN, Thompson VP, Opdam NJ, et al. Fatigue resistance of CAD/CAM resin composite molar crowns. Dent Mater 2016;32:499–509. [50] Magne P, Carvalho AO, Bruzi G, Giannini M. Fatigue resistance of ultrathin CAD/CAM complete crowns with a simplified cementation process. J Prosthet Dent 2015;114:574–9. [51] Carvalho AO, Bruzi G, Giannini M, Magne P. Fatigue resistance of CAD/CAM complete crowns with a simplified cementation process. J Prosthet Dent 2014;111:310–7. [52] Bacchi A, Dobson A, Ferracane JL, Consani R, Pfeifer CS. Thio-urethanes improve properties of dual-cured composite cements. J Dent Res 2014;93:1320–5. [53] Thomaidis S, Kakaboura A, Mueller WD, Zinelis S. Mechanical properties of contemporary composite resins and their interralations. Dent Mater 2013;29:132–41.

Please cite this article in press as: Rocca GT, et al. Fatigue behavior of resin-modified monolithic CAD–CAM RNC crowns and endocrowns. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.024