Author’s Accepted Manuscript

Fracture resistance of ceramic and polymer-based

occlusal veneer restorations

Majed Al-Akhali, Mohamed Sad Chaar, Adham

Elsayed, Abdulaziz Samran, Matthias Kern

www.elsevier.com/locate/jmbbm

PII: S1751-6161(17)30251-5
DOI: http://dx.doi.org/10.1016/j.jmbbm.2017.06.013
Reference: JMBBM2373
To appear in: Journal of the Mechanical Behavior of Biomedical Materials
Received date: 18 March 2017
Revised date: 5 June 2017
Accepted date: 9 June 2017
Cite this article as: Majed Al-Akhali, Mohamed Sad Chaar, Adham Elsayed,
Abdulaziz Samran and Matthias Kern, Fracture resistance of ceramic and
polymer-based occlusal veneer restorations, Journal of the Mechanical Behavior
of Biomedical Materials, http://dx.doi.org/10.1016/j.jmbbm.2017.06.013
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
Fracture resistance of ceramic and polymer-based occlusal veneer

restorations

Majed Al-Akhalia,b1, Mohamed Sad Chaara1, Adham Elsayeda, Abdulaziz Samrana,b,c, Matthias
Kerna
a
Department of Prosthodontics, Propaedeutics and Dental Materials, School of Dentistry,
Christian-Albrechts University, Kiel, Germany

b
Department of prosthodontics, School of Dentistry, Ibb University, Ibb, Yemen

c
Department of Restorative and Prosthetic Dental Sciences, School of Dentistry, Dar Al-Uloom
University, Riyadh, Saudi Arabia

malakhali@proth.uni-kiel.de
schaar@proth.uni-kiel.de
*
Corresponding authors at: Department of Prosthodontics, Propaedeutics and Dental Materials,
School of Dentistry, Christian-Albrechts University, Arnold- Heller-Str. 16, 24105 Kiel,
Germany. Tel.: +49 431 50026440; fax: +49 431 50026404.

ABSTRACT

OBJECTIVES

The purpose of this in vitro study was to evaluate the influence of thermodynamic loading on the

durability and fracture resistance behavior of occlusal veneers fabricated from different

biomedical dental CAD/CAM materials.

METHODS

The occlusal surfaces of 64 extracted premolars were prepared in the enamel layer and restored

with occlusal veneers with a fissure/cusp thickness of 0.5/0.8 mm made from four different

1
The first and second author contributed equally to this work.
1
dental CAD/CAM materials: group LD lithium disilicate (e.max CAD), group LS zirconia-

reinforced lithium silicate (Vita Suprinity), group PI polymer-infiltrated ceramic (Vita Enamic),

and group PM polymethylmethacrylate PMMA (Telio CAD). The prepared teeth were etched

with phosphoric acid. The occlusal veneers were then bonded using an adhesive luting system

(Multilink Primer A/B and Multilink Automix luting resin). Half of the specimens were

subjected to thermodynamic loading in a chewing simulator (1.2 million cycles at 98 N). All

specimens were quasi-statically loaded until fracture. The statistical analysis was made using the

t test and one-way ANOVA followed by the Tukey HSD test (α=0.05).

RESULTS

All aged specimens survived the thermodynamic loading. Thermodynamic loading significantly

raised the fracture resistance in groups LS, PI, and PM (P˂0.03). Occlusal veneers made from

lithium disilicate and zirconia-reinforced lithium silicate recorded higher fracture resistance than

those made from polymer-infiltrated ceramic and PMMA resin.

CONCLUSIONS

All tested dental CAD/CAM biomaterials exhibited a fracture resistance considerably exceeding

the average occlusal force in the posterior dentition. Therefore, they might present a viable long-

term treatment for restoring the occlusal surfaces of posterior teeth.

Keywords

All-ceramic; CAD/CAM; fracture resistance; occlusal veneer; thermodynamic loading

1. INTRODUCTION

Conservation of tooth structures is the main objective in restorative dentistry. Unfortunately, the

loss of natural occlusal contacts between maxillary and mandibular teeth may occur because of

wear, caries, or malposition. The treatment of such problems is complicated and challenging as
2
removal of undamaged hard dental tissues has to be done to accept conventional restorative

materials (Edelhoff and Sorensen, 2002; Tsitrou and van Noort, 2008). Moreover, loss of

occlusal contact may be compensated by tooth over-eruption (Tsitrou and van Noort, 2008),

which in turn restricts the thickness of the planned restorations. Minimally invasive procedures

in prosthetic dentistry aim to preserve sound tooth substance, maintain the vitality of the tooth,

and reduce postoperative sensitivity (Edelhoff and Sorensen, 2002).

Lithium disilicate ceramic has been considered the strongest glass-ceramic. The high number of

microstructural, interlocking, needle-like lithium disilicate crystals that are embedded in the

glassy matrix gives this type of ceramic higher mechanical properties than other types of glass-

based ceramic materials (Oh et al., 2000). Lithium disilicate ceramic has shown promising

results in terms of structural integrity when used in the anterior or posterior area as veneers,

inlays, onlays, crowns, partial coverage restorations, and three-unit fixed dental prostheses (Attia

and Kern, 2004a; Guess et al., 2013; Kern et al., 2012; Sasse et al., 2015). In an attempt to

improve the mechanical and optical properties of glass-ceramic restorative materials, new

generations have been developed for computer-aided design and computer-aided manufacturing

(CAD/CAM) technology. For instance, dissolving 10% zirconia into the lithium silicate glass

matrix claimed to be more translucent and stronger material than the conventional lithium

disilicate ceramic (Elsaka and Elnaghy, 2016). In another approach, a ceramic network structure

was infiltrated with a polymer material to combine the advantages of the two materials to obtain

better mechanical properties and better machinability for CAD/CAM than those of glass-

ceramics (Coldea et al., 2013). Likewise, the mechanical and physical properties of CAD/CAM

polymer restorative materials have been continually improved to be used as an alternative to

glass-ceramics, especially when thin restorations with high masticatory loads are required
3
because of their high resistance to dynamic fatigue (Schlichting et al., 2011). For that reason, the

longevity, stability, and behavior of these new biomedical dental materials should be investigated

when they are used for minimally invasive occlusal veneer restorations.

In the current study, the influence of thermodynamic loading on the durability and fracture

resistance of four different dental CAD/CAM occlusal veneer restorative biomaterials with a

fissure/cusp thickness of 0.5/0.8 mm was evaluated. The null hypotheses of this study were as

follows: (1) different dental CAD/CAM occlusal veneers will survive thermodynamic loading

equally; (2) thermodynamic loading will not affect the fracture resistance of each material itself;

and (3) no differences will be found in fracture resistance between the different dental

CAD/CAM biomaterials before and after thermodynamic loading.

2. MATERIALS AND METHODS

2.1. Tooth preparation

Sixty four intact, noncarious, unrestored human maxillary first premolars, recently extracted for

orthodontic reasons, were collected anonymously. They were cleaned of both calculus deposits

and soft tissues and then stored at room temperature in 0.1% thymol solution (Caelo, Hilden,

Germany). The teeth in this study were selected to be as similar as possible in dimension. The

mesiodistal and buccolingual as well as the buccal and lingual cusp slopes of the tooth occlusal

surfaces were measured and determined as 5.43 ±0.66 mm, 5.73 ±0.55 mm, 4.22 ±0.46 mm, and

3.18 ±0.34 mm, respectively. Hence, teeth below or above the average ±SD were excluded from

the study. The sample size of this study was justified to follow previous studies which were

carried out by our group with the same study design and sources of variations, and these studies’

sample sizes gave acceptable standard deviations and allowed the statistical differentiation of
4
evaluated factors. In addition, the chewing simulator machine accepts only 8 specimens per test

making this an ideal manageable group size (Attia and Kern, 2004a, b; Clausen et al., 2010;

Sasse et al., 2015).

All teeth were fixed within metallic brass tubes (Ø 15 mm) so that the exposed root portions

were coated 2 mm apical to the cemento-enamel junction with 0.2-mm-thick artificial

periodontal membrane made from a gum resin (Anti-Rutsch-Lack, Wenko-Wenselaar, Hilden,

Germany). Subsequently, an autopolymerizing resin (Technovit 4000, Heraeus Kulzer,

Wehrheim, Germany) was used to fix the coated roots inside the metallic brass tubes.

A handpiece attached to a custom-made paralleling machine and a 120-degree angulated adaptor

were used to prepare the teeth. The occlusal surfaces of the teeth were prepared in the enamel

layer with the following standardized preparation criteria: 120 degrees was the angle between the

buccal and lingual cusp slopes, mesiobuccal and distobuccal slopes, mesiolingual and

distolingual slopes, and finishing line and lingual cusp slope. All sharp angles were rounded

(Fig. 1).

2.2. Restoration fabrication

Impressions were made with the dual-mixed impression technique with polyvinylsiloxane

material (Virtual, Ivoclar Vivadent, Schaan, Liechtenstein). Thereafter, type IV stone (New

Fujirock, GC, Alsip, IL, USA) was used to form working dies, which were scanned using a 3D

scanner (D900 3D scanner, 3shape, Copenhagen, Denmark) to enable the design of the occlusal

veneers with thicknesses of 0.5/0.8 mm at fissures/cusps (Fig. 2). Thereafter, restorations were

designed in CAD-software (Dental Designer-Premium 2013, 3Shape) and fabricated with a

milling machine (inLab MC XL, Sirona, Bensheim, Germany) from four different CAD/CAM
5
materials: group LD lithium disilicate ceramic blocks (e.max CAD, Ivoclar Vivadent, Schaan,

Liechtenstein), group LS zirconia-reinforced lithium silicate ceramic blocks (Vita Suprinity, Vita

Zahnfabrik, Bad Säckingen, Germany), group PI polymer-infiltrated ceramic (Vita Enamic, Vita

Zahnfabrik), and group PM PMMA (Telio CAD, Ivoclar Vivadent). The restorations were

finished and glazed according to the manufacturers’ instructions.

2.3. Adhesive placement of restorations

The occlusal veneers were cleaned with 99% isopropanol in an ultrasonic cleaner for 3 min,

while the prepared teeth were cleaned with fluoride-free pumice for 15 sec and rinsed thoroughly

with water for 15 sec. A 5% hydrofluoric acid etching gel (IPS Ceramic Etching Gel, Ivoclar

Vivadent) was used to etch the intaglio surfaces of the occlusal veneers of groups LD and LS for

20 sec and group PI for 60 sec. The treated surfaces were cleaned thoroughly with water spray

for 60 sec. Oil-free compressed air was used to dry the intaglio surfaces, and then a silane-

containing primer (Monobond Plus, Ivoclar Vivadent) was applied immediately to the intaglio

surface of each occlusal veneer and left to react for 60 sec. Subsequently, any remaining excess

primer was dispersed with a stream of air. For group PM, the intaglio surfaces were airborne-

particle abraded with 50 µm Al2O3 at a pressure of 0.05 MPa, and then cleaned with 99%

isopropanol in ultrasonic cleaner for 3 min. Subsequently, a PMMA primer (Luxatemp-Glaze &

Bond, DMG, Hamburg, Germany) was applied, and then exposed to curing light for 20 sec using

a light-curing unit (Elipar 2500, 3M ESPE).

The prepared teeth were etched with 37% phosphoric acid (Total Etch, Ivoclar Vivadent) for 30

sec, the etchant was rinsed off thoroughly with water spray for 20 sec, and the teeth were dried

with oil-free air. Immediately afterwards, they were conditioned with a tooth primer (Multilink
6
Primer A/B, Ivoclar Vivadent), which was mixed for 10 sec in a ratio of 1:1 and applied for 30

sec to the enamel surface with a microbrush. After that, a gentle stream of air was applied to the

primed surface to evaporate volatiles, leaving the surface appearing glossy. Next, all restorations

were cemented adhesively to their respective prepared teeth with dual-polymerizing composite

resin cement (Multilink Automix, Ivoclar Vivadent). The restorations were placed in position

using gentle finger pressure and then a customized loading apparatus with a 1 kg weight. The

excess luting cement at the margins was removed with sponge pellets, and an air-inhibiting gel

(Liquid Strip, Ivoclar Vivadent) was applied along the margin of the cemented occlusal veneers

to prevent the formation of an oxygen inhibited unpolymerized resin layer. The luting material

was cured using a light-curing unit (Elipar 2500, 3M ESPE) at a distance of 5 mm for 20 sec

from the mesial, distal, buccal, and lingual directions, followed by additional curing with an

external curing unit (UniXS Kulzer, Heraeus Kulzer) for 90 sec. After cementation, the

specimens were stored in a water bath at 37°C for 3 days before testing. Finally, the specimens in

every group (n=16) were assigned randomly into 2 subgroups (n=8) either with or without

thermodynamic loading. The study groups were coded according to CAD/CAM materials as

shown in Table 1.

2.4. Cyclic loading fatigue

To mimic intraoral conditions and 5 years of clinical service (Kern et al., 1999; Steiner et al.,

2009), half of the specimens in each group (n=8) were fatigued thermodynamically in a dual-axis

computerized chewing simulator (Willytec, Feldkirchen-Westerham, Germany) so that they were

subjected to 1.2 million mechanical chewing cycles with simultaneous thermocycling between 5

and 55°C in distilled water with a 30 sec dwell time at each temperature with a total of 5,500
7
thermal cycles at a loading cycle frequency of 2.4 Hz. Steatite ceramic balls with a 6-mm

diameter (Hoechst Ceram Tec, Wunsiedel, Germany) were used as antagonists to simulate the

opposing teeth. A vertical load of 98 N (10 Kg) was applied with a vertical movement of 6 mm

and a descending speed of 30 mm/sec on the buccal cusp beginning 0.5 mm below the cusp tip

with a lateral sliding component of 0.3 mm towards the central fissure. After the thermodynamic

loading was complete, all specimens were inspected carefully under an LED light source and an

optical microscope (Wild M420, Wild Heerbrugg, Gais, Switzerland) with 5.8× magnification to

detect incipient cracks. Finally, all non-aged and surviving aged specimens were quasi-statically

loaded until fracture.

To determine the specimens’ fracture resistance, a stainless steel bar with a 6-mm-diameter ball-

end mounted in a universal testing machine (Zwick Z010/TN2A, Zwick, Ulm, Germany) was

used to apply a quasi-statical load, which was centered at the fissure along the long axis of the

restored tooth at a cross-head speed of 1 mm/min until fracture (Fig. 3). Additionally 0.5-mm tin

foil was placed between the specimens and the stainless steel bar to avoid local stress

concentration. The fracture loads were recorded automatically in Newtons (N) for each specimen

by means of testing software (testXpert II V 3.3, Zwick/Roell).

After the fracture resistance test, all specimens were inspected under an LED light source and an

optical microscope (Wild M420, Wild Heerbrugg) with 20× magnifications to evaluate the mode

of failure. According to Guess et al. (2013), failure mode was classified into four categories in

accordance with the following criteria: (I) Extensive crack formation within the restoration; (II)

cohesive fracture within the restoration; (III) adhesive fracture between the restoration and tooth

structures; (IV) longitudinal fracture of the restoration and tooth (Fig. 4).

8
2.5. Statistical analysis

One-way analysis of variance (ANOVA) was performed to compare the fracture resistance

means among the four groups, followed by multiple comparisons using the Tukey HSD test,

while the t test was used to identify which pairs of groups demonstrated a significant difference

(α=0.05). Statistical analysis was performed using SPSS 20 (IBM SPSS for Windows, Version

20.0, IBM SPSS, Inc., Chicago, IL).

3. RESULTS

All aged specimens in all groups survived the thermodynamic loading without fracture, ceramic

chipping, or cracks. The results of the quasi-static load to fracture test of all groups are listed in

Table 1. Thermodynamic loading increased the fracture resistance significantly for groups LS,

PI, and PM (P≤0.031).

Without thermodynamic loading, group LD showed significantly higher fracture resistance than

groups PI and PM (P≤0.015) but not group LS (P=0.051). No statistically significant difference

was found between groups LS, PI, PM when compared with each other (P=1.0). However, after

thermodynamic loading, group LS demonstrated significantly higher fracture resistance than

groups PI and PM (P≤0.021). In addition, group LD showed significantly higher fracture

strengths than group PM (P≤0.045). After thermodynamic loading, no statistically significant

difference was found between groups LS-LD and LD-PI (P≤0.291).

In the current study, the failure mode after quasi-static loading was evaluated and is represented

descriptively in Figure 5. It was remarkable that the most common failure modes were class I

and III for all studied groups.

9
4. DISCUSSION

Little information is available on the reliability and longevity of recent dental CAD/CAM

biomaterials as minimally invasive restorations when bonded to natural tooth structures with

nonretentive preparation designs (Clausen et al., 2010; Guess et al., 2013; Sasse et al., 2015). To

ensure a close simulation of the clinical situations, all procedures in the recent study were

designed to follow clinically applied protocols. In fact, several factors influence the fracture

resistance of all-ceramic restorations, such as microstructure, dynamic loading, fabrication

technique, final preparation design, and luting method (Kern et al., 1994). Natural human teeth

were used in this study to improve clinical relevance. Moreover, physiological tooth mobility

was simulated through the application of 0.2-mm-thick gum resin on the roots of the teeth to

replicate the periodontal membrane, which is important for the absorption and distribution of

stresses generated by masticatory forces over teeth into the alveolar bone (Poiate et al., 2009).

In vitro investigations, including thermocycling and dynamic loading are of great importance in

the testing of new dental materials with accelerated conditions mimicking the actual intraoral

activities. The parameters used for masticatory simulation were adjusted to the reported

physiological values (Kern et al., 1999; Steiner et al., 2009). Tooth morphology and preparation

geometry have been shown to influence the longevity and reliability of prosthetic restorations

(Schmidt et al., 2011). Therefore, premolars used in this current study were selected to be as

similar as possible in dimensions, i.e. mesiodistally and buccolingually. Also, the preparation

design was chosen based on the general guidelines recommended for minimally invasive partial

coverage ceramic restorations (Ahlers et al., 2009; Kern et al., 2015). To standardize the

preparation geometry, the cusp inclinations were designed to have a 120-degree angle with

rounded angles (Preis et al., 2014; Sornsuwan et al., 2011).

10
The significant reduction of occlusal veneer thickness in this study compared to the

manufacturers’ recommendations was carried out to evaluate a minimally invasive preparation of

dental hard tissues. Moreover, minimally invasive restorations provide a better bond strength

when adhesively bonded to enamel than when bonded to dentin (Ozturk et al., 2013). Besides,

thin occlusal veneer restorations bonded to enamel revealed a fracture resistance comparable to

that of thick occlusal veneers when bonded to dentin (Clausen et al., 2010; Guess et al., 2013).

The present study showed that the average final fracture resistance of all groups was higher than

that recorded for the natural unrestored human maxillary premolars, 932 N (Attia and Kern,

2004a, b). Moreover, they were higher than the maximum parafunctional masticatory forces up

to 1,000 N (Floystrand et al., 1982). The fracture resistance of the four tested dental CAD/CAM

biomaterials in this study did not correspond to their respective uniaxial flexural strength of 530,

420, 160, and 130 MPa for lithium disilicate, zirconia-reinforced lithium silicate, polymer-

infiltrated ceramic, and PMMA, respectively. This confirmed that the mechanical behavior of the

restored tooth complex, i.e. restorative material, adhesive system, and restored tooth, cannot be

predicted (Kelly, 1995; Ma et al., 2013). Furthermore, strong adhesive bonding with luting resin

can noticeably strengthen weaker ceramic restorations and balance the inherent strength

variations among different materials (Bindl et al., 2006).

No specimens in the present study showed signs of cracks or chipping after thermodynamic

loading. Thus, the first null hypothesis that different dental CAD/CAM occlusal veneers will

survive thermodynamic loading equally was accepted. Sasse et al., (2015) demonstrated that the

surviving occlusal veneers with a fissure/cusp thicknesses of 0.5/0.8 mm luted to enamel showed

cracks after thermodynamic loading. This might be attributed to the use of only self-etching

primer rather than using separate phosphoric acid etching on enamel. Self-etching primer has a
11
lower ability to etch enamel than etching with phosphoric acid. The proper morphological etch

pattern and the positive adhesion effect after using phosphoric acid on enamel has been observed

in previous studies (Erickson et al., 2009; Sheets et al., 2012; Van Landuyt et al., 2006).

In the present study, the fracture resistance was lower than that reported by Yildiz et al. (2013),

who evaluated the fracture resistance of partial coverage crowns made from machinable lithium

disilicate ceramic with an occlusal thickness of 1.5 mm, and they recorded a fracture resistance

of 2,356 ±677 N. Furthermore, Clausen et al. (2010) reported a fracture resistance of 4,070 ±777

N for occlusal veneers with a fissure/cusp thickness of 1.5/2.0 mm. Likewise, Sasse et al. (2015)

showed that occlusal veneers with thicknesses of 0.5 mm at the fissures and 0.8 mm at the cusps

had a median fracture resistance of 2,355 N. The higher overall values of fracture resistance in

the studies of Clausen et al. and Sasse et al. might be related to the less steeply prepared molar

cusp inclination angles of 150 degrees; the prepared premolar cusp inclination angle in our study

was 120 degrees. Previous studies have revealed that the failure load increases when the

prepared cusp angles increase, i.e., the less steep prepared cusp inclinations (Preis et al., 2014;

Sornsuwan et al., 2011). Furthermore, differences in preparation design, ceramic thicknesses,

tooth morphology, and study design might have influenced the fracture resistance.

Guess et al. (2013), measured the fracture resistance of retentive occlusal veneers made from

pressed lithium disilicate. Their occlusal onlays with thicknesses of 0.5, 1.0, and 2.0 mm had a

lower fracture resistance of 997 ±331, 1108 ±340, and 963 ±405 N, respectively, after

thermodynamic loading than the CAD/CAM lithium disilicate occlusal veneers in the current

study with 1,545 ±175 N. The lower mean of fracture resistance in the former study might be due

to the extension of the occlusal onlays to restore the occlusal and the proximal walls of the

premolar teeth. Guess et al. emphasized that occlusal veneers with a 0.5 mm thickness bonded to
12
enamel had a fracture resistance comparable with that of 2.0 mm thick occlusal veneers bonded

to dentin. The improved fracture resistance of the ultrathin occlusal onlay is due to the superior

bonding to enamel as compared when bonding to dentin.

Thermodynamic loading significantly raised the fracture resistance of groups LS, PI, and PM in

the present study. Therefore, the second null hypothesis that thermodynamic loading will not

affect the fracture resistance of each material was partially rejected. This finding might be

explained by the postcuring effect and stress relaxation because of the warm temperature of the

thermocycling water (Par et al., 2014; Vouvoudi and Sideridou, 2013).

Regarding the comparison among the different dental CAD/CAM biomaterials without

thermodynamic loading, lithium disilicate occlusal veneers demonstrated significantly higher

fracture resistance than groups PI and PM (P≤0.015). As a result, the first part of the third null

hypothesis that there will be no differences in fracture resistance among the different dental

CAD/CAM materials before thermodynamic loading was partially rejected. This might be

attributed to the higher mechanical properties of lithium disilicate ceramic (Bindl et al., 2006) or

it could be related to the monolithic structural property of lithium disilicate ceramic which

facilitates a proper etching pattern using hydrofluoric acid, so that a stronger bond with the

adhesive resin cement may be achieved (Borges et al., 2003).

However, after thermodynamic loading, occlusal veneers made from zirconia-reinforced lithium

silicate (group LS) showed significantly higher fracture resistance than occlusal veneers made

from the resin-containing materials (groups PI and PM). Lithium disilicate (group LD) showed

significantly higher fracture resistance than PMMA restorations (group PM). Therefore, the

second part of the third null hypothesis that no differences will be found in fracture resistance

among the different dental CAD/CAM materials after thermodynamic loading was also partially
13
rejected. The reason for this result might be the inherent mechanical properties of the tested

restorative materials (Awada and Nathanson, 2015; Bindl et al., 2006; Elsaka and Elnaghy,

2016). In some studies, CAD/CAM thin composite resin occlusal veneers had significantly high

stepwise loading fatigue resistance when compared with ceramic veneers (Magne et al., 2010;

Schlichting et al., 2011). However, the loading protocols used in these studies were quite

different from that of the current study, as they followed a stepwise loading fatigue from 200 N

up to 1,400 N at a maximum of 185,000 cycles without testing the specimens under static load,

which limits the comparability.

5. CONCLUSION

Considering the survivability and the fracture resistance of the tested occlusal veneers, all

materials may be considered as a viable treatment for restoring the occlusal surfaces of posterior

teeth.

Acknowledgements

The authors gratefully acknowledge Ivoclar Vivadent, Schaan, Liechtenstein, and Vita

Zahnfabrik, Bad Säckingen, Germany for providing their materials free of charge.

REFERENCES

Ahlers, M.O., Morig, G., Blunck, U., Hajto, J., Probster, L., Frankenberger, R., 2009. Guidelines

for the preparation of CAD/CAM ceramic inlays and partial crowns. Int. J. Comput. Dent. 12,

309-325.

14
Attia, A., Kern, M., 2004a. Fracture strength of all-ceramic crowns luted using two bonding

methods. J. Prosthet. Dent. 91, 247-252.

Attia, A., Kern, M., 2004b. Influence of cyclic loading and luting agents on the fracture load of

two all-ceramic crown systems. J. Prosthet. Dent. 92, 551-556.

Awada, A., Nathanson, D., 2015. Mechanical properties of resin-ceramic CAD/CAM restorative

materials. J. Prosthet. Dent. 114, 587-593.

Bindl, A., Luthy, H., Mormann, W.H., 2006. Strength and fracture pattern of monolithic

CAD/CAM-generated posterior crowns. Dent. Mater. 22, 29-36.

Borges, G.A., Sophr, A.M., de Goes, M.F., Sobrinho, L.C., Chan, D.C., 2003. Effect of etching

and airborne particle abrasion on the microstructure of different dental ceramics. J. Prosthet.

Dent. 89, 479-488.

Clausen, J.O., Abou Tara, M., Kern, M., 2010. Dynamic fatigue and fracture resistance of non-

retentive all-ceramic full-coverage molar restorations. Influence of ceramic material and

preparation design. Dent. Mater. 26, 533-538.

Coldea, A., Swain, M.V., Thiel, N., 2013. Mechanical properties of polymer-infiltrated-ceramic-

network materials. Dent. Mater. 29, 419-426.

Edelhoff, D., Sorensen, J.A., 2002. Tooth structure removal associated with various preparation

designs for posterior teeth. Int. J. Periodontics Restorative Dent. 22, 241-249.

Elsaka, S.E., Elnaghy, A.M., 2016. Mechanical properties of zirconia reinforced lithium silicate

glass-ceramic. Dent. Mater. 32, 908-914.

Erickson, R.L., Barkmeier, W.W., Kimmes, N.S., 2009. Bond strength of self-etch adhesives to

pre-etched enamel. Dent. Mater. 25, 1187-1194.

15
Floystrand, F., Kleven, E., Oilo, G., 1982. A novel miniature bite force recorder and its clinical

application. Acta Odontol. Scand. 40, 209-214.

Guess, P.C., Schultheis, S., Wolkewitz, M., Zhang, Y., Strub, J.R., 2013. Influence of

preparation design and ceramic thicknesses on fracture resistance and failure modes of premolar

partial coverage restorations. J. Prosthet. Dent. 110, 264-273.

Kelly, J.R., 1995. Perspectives on strength. Dent. Mater. 11, 103-110.

Kern, M., Beuer, F., Frankenberger, R., Kohal, R., Kunzelmann, K., Mehl, A., Pospiech, P.,

Reiss, B., Wiedhahn, K., 2015. All-Ceramic-at a glance, 6th ed. AG für Keramik in der

Zahnheilkunde eV, Ettlingen.

Kern, M., Fechtig, T., Strub, J.R., 1994. Influence of water storage and thermal cycling on the

fracture strength of all-porcelain, resin-bonded fixed partial dentures. J. Prosthet. Dent. 71, 251-

256.

Kern, M., Sasse, M., Wolfart, S., 2012. Ten-year outcome of three-unit fixed dental prostheses

made from monolithic lithium disilicate ceramic. JADA 143, 234-240.

Kern, M., Strub, J.R., Lu, X.Y., 1999. Wear of composite resin veneering materials in a dual-axis

chewing simulator. J. Oral Rehabil. 26, 372-378.

Ma, L., Guess, P.C., Zhang, Y., 2013. Load-bearing properties of minimal-invasive monolithic

lithium disilicate and zirconia occlusal onlays: finite element and theoretical analyses. Dent.

Mater. 29, 742-751.

Magne, P., Schlichting, L.H., Maia, H.P., Baratieri, L.N., 2010. In vitro fatigue resistance of

CAD/CAM composite resin and ceramic posterior occlusal veneers. J. Prosthet. Dent. 104, 149-

157.

16
Oh, S.C., Dong, J.K., Luthy, H., Scharer, P., 2000. Strength and microstructure of IPS Empress 2

glass-ceramic after different treatments. Int. J. Prosthodont. 13, 468-472.

Ozturk, E., Bolay, S., Hickel, R., Ilie, N., 2013. Shear bond strength of porcelain laminate

veneers to enamel, dentine and enamel-dentine complex bonded with different adhesive luting

systems. J. Dent. 41, 97-105.

Par, M., Gamulin, O., Marovic, D., Klaric, E., Tarle, Z., 2014. Effect of temperature on post-cure

polymerization of bulk-fill composites. J. Dent. 42, 1255-1260.

Poiate, I.A., de Vasconcellos, A.B., de Santana, R.B., Poiate, E., 2009. Three-dimensional stress

distribution in the human periodontal ligament in masticatory, parafunctional, and trauma loads:

finite element analysis. J. Periodontol. 80, 1859-1867.

Preis, V., Dowerk, T., Behr, M., Kolbeck, C., Rosentritt, M., 2014. Influence of cusp inclination

and curvature on the in vitro failure and fracture resistance of veneered zirconia crowns. Clin.

Oral Investig. 18, 891-900.

Sasse, M., Krummel, A., Klosa, K., Kern, M., 2015. Influence of restoration thickness and dental

bonding surface on the fracture resistance of full-coverage occlusal veneers made from lithium

disilicate ceramic. Dent. Mater. 31, 907-915.

Schlichting, L.H., Maia, H.P., Baratieri, L.N., Magne, P., 2011. Novel-design ultra-thin

CAD/CAM composite resin and ceramic occlusal veneers for the treatment of severe dental

erosion. J. Prosthet. Dent. 105, 217-226.

Schmidt, K.K., Chiayabutr, Y., Phillips, K.M., Kois, J.C., 2011. Influence of preparation design

and existing condition of tooth structure on load to failure of ceramic laminate veneers. J.

Prosthet. Dent. 105, 374-382.

17
Sheets, J.L., Wilcox, C.W., Barkmeier, W.W., Nunn, M.E., 2012. The effect of phosphoric acid

pre-etching and thermocycling on self-etching adhesive enamel bonding. J. Prosthet. Dent. 107,

102-108.

Sornsuwan, T., Ellakwa, A., Swain, M.V., 2011. Occlusal geometrical considerations in all-

ceramic pre-molar crown failure testing. Dent. Mater. 27, 1127-1134.

Steiner, M., Mitsias, M.E., Ludwig, K., Kern, M., 2009. In vitro evaluation of a mechanical

testing chewing simulator. Dent. Mater. 25, 494-499.

Tsitrou, E.A., van Noort, R., 2008. Minimal preparation designs for single posterior indirect

prostheses with the use of the Cerec system. Int. J. Comput. Dent. 11, 227-240.

Van Landuyt, K.L., Kanumilli, P., De Munck, J., Peumans, M., Lambrechts, P., Van Meerbeek,

B., 2006. Bond strength of a mild self-etch adhesive with and without prior acid-etching. J. Dent.

34, 77-85.

Vouvoudi, E.C., Sideridou, I.D., 2013. Effect of food/oral-simulating liquids on dynamic

mechanical thermal properties of dental nanohybrid light-cured resin composites. Dent. Mater.

29, 842-850.

Yildiz, C., Vanlioglu, B.A., Evren, B., Uludamar, A., Kulak-Ozkan, Y., 2013. Fracture resistance

of manually and CAD/CAM manufactured ceramic onlays. J. Prosthodont. 22, 537-542.

18
Figures:

Figure 1. Occlusal surface preparation with 120 degrees between the prepared surfaces.

Figure 2. A virtual design of an occlusal veneer for one of the studied specimens.

19
Figure 3. Load to fracture test in the universal testing machine (Zwick) before the placement

of the tin foil.

20
Figure 4. Failed specimens in occlusal view showing the failure mode: I=Extensive crack

formation within restoration. II=Cohesive fracture within restoration. III=Fracture within

restoration and tooth structures. IV=Longitudinal restoration and tooth fracture involving root.

10 9 6
15
41 26
32 50 57 57
31 47

12 Class IV
16 30
12 23 12 Class III
Class II
42 47 43 43
38 35 31 35 Class I

LD LS PI PM LD LS PI PM
Without thermodynamic loading With thermodynamic loading

Figure 5. Descriptive statistics of failure mode (%). Definition of failure modes class I to IV see

Fig. 4.

21
Table 1 - Fracture resistance of groups in Newtons [N], means, standard deviations (SD), medians,

lower and upper quartiles, minima and maxima (n=8).

Without thermomechanical loading After thermomechanical loading

Groups (Codes) Mean ±SD Median Q1 Min Mean ±SD Median Q1 Min

Q3 Max Q3 Max

Lithium disilicate 1,408.8 1,335.0 1,220.0 1,150.0 1,545.0 1,560.0 1,372.5 1,300.0

(LD) ±215.8 Aa 1,635.0 1,730.0 ±175.2 ABa 1,700.0 1,770.0

Zirconia-reinforced 1,076.8 1,015.0 839.5 616.0 1,667.5 1,735.0 1,495.0 1,360.0

lithium silicate ±324.9 ABa 1,285.0 1,670.0 ±189.1 Ab 1,842.5 1,860.0

(LS)

Polymer-infiltrated 1,018.5 1,005.0 863.5 798.0 1,321.0 1,310.0 1,162.5 848.0

ceramic ±155.5 Ba 1,185.0 1,190.0 ±269.1 BCb 1,567.5 1,680.0

(PI)

PMMA 974.5 994.0 845.5 593.0 1,232.5 1,130.0 1,032.5 1,010.0

(PM) ±208.4 Ba 1,112.8 1,290.0 ±223.1 Cb 1,497.5 1,530.0

Means with the same upper case superscript letters within the same column are not statistically different (P˃0.05).

Means with the same subscript letters within the same raw are not statistically different (P˃0.05).

22