Fi

CDA

a

A

R

R

2

A

K

P

F

F

Z

Y

I

C

1

Aetfip

0d

d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 67–75

avai lab le at www.sc iencedi rec t .com

journa l homepage: www. int l .e lsev ierhea l th .com/ journa ls /dema

racture strength of prefabricated all-ceramic posteriornlay-retained fixed dental prostheses

hristian Mehl ∗, Klaus Ludwig, Martin Steiner, Matthias Kernepartment of Prosthodontics, Propaedeutics and Dental Materials, Dental School, Christian-Albrechts University Kiel,rnold-Heller-Straße 16, 24105 Kiel, Germany

r t i c l e i n f o

rticle history:

eceived 19 December 2008

eceived in revised form

3 July 2009

ccepted 29 July 2009

eywords:

refabricated restorations

DP

racture load

irconia ceramic

-TZP

nlay-retained

omposite

a b s t r a c t

Objectives. The purpose of this in vitro study was to compare the centric and eccentric

quasi-static and fatigue fracture strength of industrially prefabricated resin-bonded three-

unit inlay-retained fixed dental prostheses (IPIRFDPs). The IPIRFDPs consisted of industrial

manufactured yttria-stabilized tetragonal zirconia (Y-TZP) frameworks with an industrially

added microhybrid composite veneering.

Methods. Identical IPIRFDP-models consisted of a second premolar, a missing first molar

and a second molar (CoCrMo alloy) integrated in a low melting alloy base. Roots were

covered with a soft silicone layer to simulate an artificial parodontium. Premolars had an

occlusal–distal inlay-preparation and molars a mesial–occlusal inlay-preparation. Forty-two

IPIRFDPs with a connector size of 9 mm2 and a framework connector size of 4.7 mm2 were

cemented adhesively to the IPIRFDP-models. Quasi-static fracture strength was tested with

centric (n = 12) and eccentric (n = 6) loading in a universal testing machine at a cross-head

speed of 1 mm/min. Fatigue fracture strength was tested at 1200 N with centric loading

(n = 12) and at 600/500 N with eccentric loading (n = 6) at a frequency of 0.5 Hz. Statistical

comparison of groups was performed with the Mann–Whitney U test.

Results. Quasi-static fracture strength differed significantly between centric (1749 N) and

eccentric loading (880 N, p < 0.001). Mean loading cycles until fracture were 4432 for centric

loading at 1200 N compared to only 3 and 410 loading cycles for eccentric loading at 600 and

500 N, respectively.

Significance. Considering the maximum chewing forces in the molar region, it seems clini-

cally possible to use prefabricated IPIRFDPs with Y-TZP as a core material with a framework

connector size of 4.7 mm2.

emy

spective of the type of FDP (porcelain fused to metal vs.

© 2009 Acad

. Introduction

mong the permanent dentition, the first molar has the high-st incidence of tooth loss due to caries, endodontic reasons,

ooth fractures and periodontal diseases [1,2]. However therst molar has a significant role in maintaining the intercus-al and condylar position [3,4]. Therefore the replacement of a

∗ Corresponding author at: Private Practice, 10 Brook Street, W1S 1BG LonE-mail address: cmehl@proth.uni-kiel.de (C. Mehl).

109-5641/$ – see front matter © 2009 Academy of Dental Materials. Puoi:10.1016/j.dental.2009.07.013

of Dental Materials. Published by Elsevier Ltd. All rights reserved.

missing molar might be an important consideration to preventfurther functional problems [2,4].

The traditional way of molar replacement is either a fixeddental prosthesis (FDP) or an implant retained crown [5]. Irre-

don, United Kingdom. Tel.: +44 845 0944004; fax: +44 845 0943003.

all-ceramic crown) the clinician uses, a crown preparation isalways a risk to pulp vitality and may lead to pulpal reactionsin the long term [6]. Approximately 63–73% of the coronal tooth

blished by Elsevier Ltd. All rights reserved.

68 d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 67–75

Table 1 – Materials used in this study.

Product Material type Manufacturer Batch no.

Gapless hydrofluoric etchant Hydrofluoric etchant Gapless GmbH, Umkirch, Germany LOT 07161Calibra Silane Coupling Agent Silane Dentsply DeTrey, Konstanz, Germany LOT 070511XP-Bond Bonding agent Dentsply DeTrey, Konstanz, Germany LOT 0702001786SCA Self-curing activator Dentsply DeTrey, Konstanz, Germany LOT 070119Calibra esthetic resin cement Self-curing resin based

dental luting cementDentsply DeTrey, Konstanz, Germany LOT 070525 (base)

Base/Catalyst LOT 0705291 (cat.)Airblock Oxygen blocking gel Dentsply DeTrey, Konstanz, Germany LOT 072161

and 2.3 mm below the framework allows the dentist to adjustthe base and the occlusal area of the pontic. The veneer-ing microhybrid composite contains 24% (weight) organicmatrix (Bis-GMA and Diurethandimethacylat) and 75% silane

Fig. 1 – The industrial prefabricated three-unitinlay-retained FDP with a framework made ofyttria-stabilized tetragonal zirconia and microhybrid

Gapless FPD Prefabricated all-ceramicinlay-retained three-unitFPDs

structure is removed when teeth are prepared for crowns [7].Regarding these facts it seems desirable to adapt the type ofabutment preparation to the extent of sound tooth structureafter caries removal not only for a single tooth restoration, butalso for abutment preparations for FDPs. Therefore, if a patientrejects an implant treatment and enough sound tooth struc-ture is available it would be desirable to restore a missing toothwith an inlay-retained FDP instead of a crown-retained FDP.

Clinical evaluations of inlay-retained three-unit FDPsshowed a failure rate of 10% [8] after 9 months (Empress II)and 13% [9] after 37 months (IPS e.max Press, Ivoclar-Vivadent,Schaan, Liechtenstein). In both studies the failure was evokedby debonding or a combination of both debonding and frac-ture. Despite these failure rates the inlay-retained FDP couldbe a favorable treatment option with respect to biological [6,7]and economic reasons [5].

Recently, yttria-stabilized tetragonal zirconia (Y-TZP)has been made available to dentistry through CAD/CAM-techniques [10] and provides excellent mechanical perfor-mance, superior strength and fracture resistance comparedto other ceramics [11]. Therefore Y-TZP ceramic might be analternative material for the fabrication of inlay-retained FDPsto minimize the risk of fracture.

In the present study the quasi-static and fatigue fracturestrengths of industrially prefabricated inlay-retained FDPs,consisting of Y-TZP frameworks with a microhybrid compos-ite veneering, were determined. The null hypothesis testedwas that centric and eccentric loading does not influence theirfracture strength. In addition, it should be evaluated whetherthese FPDs provide fracture strengths, which make them aviable treatment option to replace a missing first molar duringa single patient appointment.

2. Materials and methods

All materials used and batch numbers are summarized inTable 1.

2.1. Prefabricated all-ceramic three-unit inlay-retainedfixed dental prostheses

In this study industrially prefabricated three-unit inlay-retained fixed dental prostheses (IPIRFDPs) were used(Gapless, Umkirch, Germany, Fig. 1). They are designed torestore upper or lower first molars chairside in one session.

Gapless GmbH, Umkirch, Germany LOT 10084-1-2+3/88,LOT 10084-1-2+3/89

Three sizes (9, 10, and 11 mm pontic size) and three differentcolors (A, B, C, Gapless colors) are available.

The IPIRFDPs consist of an industrially manufacturedyttria-stabilized tetragonal zirconia (Y-TZP) framework, athin layer of feldspathic ceramic and microhybrid compositeveneering [12]. The Y-TZP framework has a connector heightof 2.5 mm (4.7 mm2) and provides a high fracture strength ofIPIRFDPs, which was in previous tests 1200 N [13]. A thin layerof feldspathic ceramic is fired to the framework in order toincrease the bond strength between the Y-TZP framework andthe veneering microhybrid composite and to enable the den-tist to etch the surface of the connector area with hydrofluoricacid in preparation for the adhesive cement.

The microhybrid composite veneering is added industri-ally to the framework. The composite has a connector heightof 2.3 mm above the Y-TZP framework, which leads to an over-all connector height of 4.8 mm (9 mm2). The height of thecomposite veneering in the pontic region of 2.9 mm above

composite veneering. The height of the veneeringmicrohybrid composite is 2.9 mm above and 2.3 mm belowthe framework, which leads to an overall connector size of9 and 4.7 mm2 framework connector size.

s 2 6 ( 2 0 1 0 ) 67–75 69

c0ih9Tp9a

2

Aism(((Ct(ttTagm[pHid

Fig. 2 – Schematic diagram of a test specimen. Abutmentswere made from a cobalt-chromium alloy. Roots werecovered with a 200 �m thick soft silicone layer. The basisaround the teeth was made from a low melting alloy (MCP

F(dwc

d e n t a l m a t e r i a l

oated, inorganic siliciumdioxid filler particles, with a size of.04–3 �m and an average filler particle size of 0.7 �m. Accord-ng to the manufacturer, the microhybrid composite resinas a fracture strength of 145 MPa and an elastic modulus of700 MPa (three-point flexural strength (DIN 13922/EN 24049)).he shear bond strength of the veneering microhybrid com-osite to the feldspathic layer on the Y-TZP framework was00 N [14]. The IPIRFDPs have been used as delivered withoutny modification.

.2. Master models

rtificial teeth (No. 15 and 17, KaVo, Biberach, Germany) werentegrated in an otherwise toothless KaVo model. A box-haped inlay-preparation (3◦ divergence, 4.8 mm height, 1 mminimal depth) was performed with a special diamond bur

100 �m grained) according to the manufacturer’s instructionsGapless preparation set, Gapless). The teeth were duplicatedSpeedy-duplication-technique [15]) and afterwards cast in aoCrMo alloy (Wironit, Bego, Bremen, Germany). To create

he master models one IPIRFDP was cemented provisionallyFreegenol, GC, Tokyo, Japan) to the prepared artificial teeth inhe KaVo model and teeth and IPIRFDP afterwards duplicatedogether (Transpaduplisil, Wichnalek, Augsburg, Germany).he roots of the manufactured metal teeth were covered with200 �m thick soft silicone layer (Erkoskin, Erkodent, Pfalz-

rafenweiler, Germany) to provide an artificial periodontalembrane, which allowed a periodontal movement of 50 �m

16]. Tooth mobility was controlled with an electronic dis-

lacement transducer under a horizontal load of 20 N (W1T3,BM, Darmstadt, Germany). Next, the prepared teeth were

nserted into the duplicated form of the master models and theuplicated form filled with a low melting alloy (MCP 70, HEK,

ig. 3 – Workflow from master model to specimen: (A) PreparatioB) Both teeth were duplicated in cobalt-chromium alloy using aies were covered with an artificial periodontal membrane, theyith a low melting alloy. (D) Specimen removed from the duplica

ementation.

70). The dimensions (mm) of the prefabricatedinlays-retained FDP are shown.

Lübeck, Germany). Fig. 2 exemplarily shows the schematic dia-gram of one specimen (IPIRFDP-model) and Fig. 3 explains the

workflow from master model to specimen.

According to this procedure, 42 identical specimens withresilient supported teeth were fabricated and divided into thefollowing four groups: centric (group CQSFS) and eccentric

n of the inlay cavities with the preparation guide (Gapless).wax duplication technique. (C) After the roots of the metalwere inserted in a duplication form and the form was filledtion form and after cleaning procedure ready for adhesive

l s 2

70 d e n t a l m a t e r i a

(group EQSFS) quasi-static fracture strength testing and cen-tric (group CFFS) and eccentric (group EFFS) fatigue fracturestrength testing.

2.3. Adhesive cementation

For all groups the surfaces of the IPIRFDPs were conditionedby etching with 5% hydrofluoric acid (Gapless) for 30 s and anapplication of silane coating (Calibra Silane, Dentsply DeTrey,Constance, Germany) for 1 min. The inlay cavities of the metalabutments were air-abraded with 110 �m alumina particlesfor 10 s at 0.3 MPa pressure. To ensure that the air-abradedsurface was free of loose alumina particles, the metal abut-ments were ultrasonically cleaned for 3 min in isopropanol(96%) [17]. In previous studies [18–20] this procedure showedno negative effect on the bond strength of the luting resin.Finally, XP-BOND and SCA (Dentsply DeTrey) were mixed (1:1),applied to the IPIRFDPs and the inlay cavities for 3 min, but notlight cured. After these conditioning procedures IPIRFDPs ofall groups were cemented to the metal abutments with com-posite resin cement (Calibra, Dentsply DeTrey) according tothe manufacturer’s recommendations. After removing excessluting resin, an air-blocking gel (Airblock, Dentsply DeTrey)was applied to the bonding margins and the luting resin wasthen light cured for 40 s at each of three different sites. Thecemented IPIRFDPs were stored in distilled water for 3 days at37 ◦C.

2.4. Testing procedures

A summary of all testing procedures and parameters is givenin Fig. 4. The centric quasi-static fracture strength (group

Fig. 4 – Flow chart describing all the tes

6 ( 2 0 1 0 ) 67–75

CQSFS, n = 16) was tested in a universal testing machine (ZwickBZD10/TNZA, Ulm, Germany). The load was applied perpen-dicular to and at the centre of the IPIRFDPs using a stainlesssteel ball (diameter 5 mm) at a cross-head speed of 1 mm/minuntil fracture occurred. To avoid primary cracks at the pointof loading a 0.5 mm foil of Makrolon (Macrolon, Bayer, Lev-erkusen, Germany) was inserted for all groups.

The eccentric quasi-static fracture strength (group EQSFS,n = 8) was also tested in the universal testing machine (ZwickBZD10/TNZA), but this time the load was applied 3 mm eccen-tric to the mesio-distal direction at the fissure between themesiobuccal and distobuccal cusp of the IPIRFDPs. A stainlesssteel stamp (diameter 1.5 mm) was used at a cross-head speedof 1 mm/min until fracture occurred.

The centric fatigue fracture strength (group CFFS, n = 12)was tested in a universal testing machine (Zwick 1435, Ulm,Germany) and the specimen subjected to a dynamic loadbetween 10 and 1200 N with a frequency of 0.5 Hz until thefirst fracture event occurred. As before, the load was appliedperpendicular to and at the centre of the IPIRFDP pontic usinga stainless steel ball (diameter 5 mm) and a 0.5 mm foil ofmacrolon. The steel ball was not lifted from the IPIRFDPs dur-ing loading. With these settings and the application of a basicloading of 10 N an impact pulse could be avoided.

The eccentric fatigue fracture strength (group EFFS, n = 6)was also tested in a universal testing machine (Zwick 1435,Ulm, Germany). The specimens were subjected to a dynamicload between 10 and 600 N with a frequency of 0.5 Hz until

the first fracture event occurred. As in group EQSFS the loadwas applied 3 mm eccentric to the mesio-distal direction atthe fissure between the mesiobuccal and distobuccal cusp ofthe IPIRFDPs using a stainless steel stamp (diameter 1.5 mm).

t procedures and their parameters.

d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 67–75 71

Table 2 – Medians of centric and eccentric quasi-static fracture strengths of groups CQSFS (n = 16) and EQSFS (n = 8) in N.

Group CQSFS Group EQSFS

Batch no. LOT 10084-1-2+3/88 LOT 10084-1-2+3/89 Pooled LOT 10084-1-2+3/89

Median 1504 1789 1749 88025th percentile 1433 1674 1477 80475th percentile 1910 1869 1891 918

entrie resu

Asr

2

Dich

3

3

Tgut

(1

Fclf

Group CQSFS, centric quasi-static fracture strength; group EQSFS, eccspecimen of the two batch numbers of group EQSFS and therefore th

t the top load of 600 N the fractures occurred for the first twopecimens after a few trials and therefore the top load waseduced to 500 N (n = 4) for group EFFS.

.5. Statistical analyses

ue to the small group sizes and abnormally distributed datan some groups non-parametric testing was calculated byomparison of medians with the Mann–Whitney U test. Allypothesis testing was conducted at a 95% level of confidence.

. Results

.1. Quasi-static fracture strength

he medians of the quasi-static fracture strengths of allroups are shown in Table 2. The fracture strength was testedntil the specimen fractured completely, previous fractures of

he veneering ceramic have not been considered.

For group CQSFS no statistical differences for the mediansp = 0.49) between the two batch numbers (batch no. 1 = LOT0084-1-2+3/88, n = 8; batch no. 2 = LOT 10084-1-2+3/89, n = 8)

ig. 5 – Selected samples of fractured IPIRFDPs. The typical fractuentric quasi-static load and (B) fatigued with a centric load. Theamella, directly followed by a central fracture of the framework.atigued with an eccentric load. When an eccentric load was app

c quasi fracture strength. No difference could be found between thelts where pooled (Mann–Whitney test, p = 0.49).

could be found and therefore the results for group CQSFS forboth batch numbers have been pooled. The median quasi-static fracture strength for group CQSFS was 1749 N and forgroup EQSFS (n = 8) 880 N. Examples for typical fractures areshown in Fig. 5A and C.

3.2. Fatigue tests

The medians and single values of loading cycles performeduntil the first fracture occurred are shown for all groups inTable 3. For group CFFS no statistical differences for the medi-ans (p = 0.69) between the two batch numbers (batch no. 1, n = 6;batch no. 2, n = 6) could be found and therefore the results forgroup CFFS have been pooled.

Two fracture events of the veneering composite occurredin group CFFS (specimens 2 and 5 of batch no. 2) before theframework fractured. Statistical testing revealed no significant

difference between framework and veneering composite frac-ture (p ≥ 0.05) and therefore all 12 single values of group CFFShave been analyzed. The mean number of loading cycles forgroup CFFS was 4432.

re lines are shown by black lines: (A) Specimen tested withveneering composite fractured first, especially the buccal(C) Specimen tested with eccentric quasi-static load and (D)lied, the connectors fractured first.

72 d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 67–75

Table 3 – Medians and single values of loading cycles for the first fracture event for groups CFFS (n = 12) and EFFS (n = 2 for600 N, n = 4 for 500 N) in N.

Dynamic load Group CFFS Group EFFS (no. 1 + 2 at 600 N, no. 2–6 at 500 N)

Batch no. LOT 10084-1-2+3/88 LOT 10084-1-2+3/89 LOT 10084-1-2+3/89

1200 N (n = 12), median 1371 2660 –No. 1 91 4976 1No. 2 14,487 30a 5No. 3 205 345 101No. 4 9230 12,069 32No. 5 2008 27a 703No. 6 735 8983 805Mean 4432b 3c

410d

Group CFFS, centric fatigue fracture strength; group EFFS, eccentric fatigue fracture strength.a Fracture of the veneering composite, no framework fracture.b Mean at a tested load of 1200 N is shown. Mean calculated for specimens of both batch numbers as no difference between the specimens of

the batch numbers was found (Mann–Whitney test, p = 0.69).

c Mean at a tested load of 600 N is shown.d Mean at a tested load of 500 N is shown.

For group EFFS the number of loading cycles for 600 N (n = 2)loading force was one and five loading cycles for specimens 1and 2 and therefore the applied loading force was reduced to500 N. Since the number of specimens tested was too small tocalculate the median, the mean of 410 loading cycles was cal-culated for group EFFS at 500 N. Examples for typical fracturesare shown in Fig. 5B and D.

4. Discussion

In this study industrially prefabricated three-unit inlay-retained fixed dental prostheses (IPIRFDPs, Gapless) wereused. They are designed to restore upper or lower firstmolars chairside in one session. Identical shape of all sam-ples assured identical stress conditions for all samples, sincedifferent geometry of inlay-retained FDPs can cause differ-ent fracture patterns [16]. Additionally, to minimize variablefactors and in an effort to get homogenous results in thepresent in vitro study, standardized “master teeth” were pre-pared with a preparation positioning guide (Gapless) andthen duplicated. Therefore the IPIRFDPs could be bonded tonearly identical metal casts using the protocol described inSection 2.

The Y-TZP frameworks of the IPIRFDPs were industriallymanufactured. Y-TZP ceramic has proved its superior materialcharacteristics in various studies [11,16]. It consists of partiallystabilized zirconia particles (95% ZrO2 partially stabilized by5% Y2O3) with a mean grain size of 0.4 �m. These particles aredensely sintered, resulting in a final microstructure in whichvoids, flaws and cracks are reduced to a minimum [21]. For thisreason and because of the transformation-toughening mech-anism Y-TZP frameworks offer remarkable fracture strengths[16,21].

The veneering of the framework was conducted industri-

ally by adding a microhybrid composite. The wear resistanceof this composite has shown to be comparable to other micro-hybrid composites [22]. Because of a reported high incidencefor chipping of veneering ceramics [23] this material might

offer an approach to reduce chipping. However, any repair oradjustment can easily be performed by adding or removingcomposite [24,25].

New dental materials in combination with new preparationdesigns have to be tested before they can be recommendedfor clinical use [16]. It is not known what fracture resistance isneeded to achieve a good long-term outcome of IPIRFDPs in themolar region. Numerous authors have investigated the maxi-mum bite forces during mastication and mean values for themaximum bite force level varied from 216 to 847 N [26–28]. Thehighest bite force was found in the first molar region. Review-ing the literature, Körber and Ludwig [29] summarized thatposterior FDPs should be strong enough to withstand a loadof 500 N.

Additionally, cyclic fatigue loading caused by masticationcan considerably weaken the fracture resistance of all-ceramicrestorations [30]. Under the conditions of the oral environ-ment, the inherent flaws of ceramic materials act as the originof crack propagation and can grow to critical sizes [21]. Theendurance limit for fatigue cycling that can be applied to den-tal ceramics is approximately 50% of the maximal fracturestrength [31]. Therefore it seems reasonable to assume thatan initial fracture resistance of 1000 N should be required fora favorable clinical prognosis of posterior inlay-retained FDPswith all-ceramic frameworks. With 1749 N the median of groupCQSFS exceeded this threshold value by far. Other studies[16,21,32] with centric Y-TZP framework testing showed frac-ture strength values of 1000–3000 N. In a comparable study[16] Y-TZP inlay-retained FDPs with a connector size of 9 mm2

showed values for centric quasi-static fracture strength test-ing of 3100 N.

However, the way of testing Y-TZP frameworks centricallyfavors the characteristics of this material [16,21] and it doesnot correspond to the reality of the oral cavity, where forces arenot always applied perpendicular to the axis of a restoration.

Therefore in group EQSFS specimens were tested eccentri-cally to the mesio-distal axis of the IPIRFDPs and showed asignificantly reduced median fracture strength of only 880 N.Therefore, the hypothesis that centric and eccentric loading

s 2 6

dIblo

IScamgtoamsTtafs

tpcsbwcmc

itliIta(tbathoeatieIimtgt

m

d e n t a l m a t e r i a l

oes not influence the fracture strength has to be rejected. SoPIRFDPs with a framework connector size of 4.7 mm2 seem toe sufficiently strong regarding the threshold value for centric

oading, but must be reviewed critically when eccentric forcesccur.

Additional to quasi-static fracture strength testing, thePIRFDPs were fatigued under centric and eccentric loading.ince the testing in a chewing simulator needs too manyycles to fatigue Y-TZP FDPs and an impact pulse should bevoided [16], specimens were tested in a universal testingachine until the first fracture occurred. The specimens in

roups CFFS/EFFS were tested at a maximum load of 70% ofhe median centric/eccentric quasi-static values. Specimensf group CFFS were fatigued at a maximum load of 1200 Nfter 4432 cycles. For eccentric fatigue tests (group EFFS) aaximum load of 600 N was chosen initially, but the first two

pecimens fractured after one and five cycles, respectively.herefore the maximum load was reduced to 500 N and due

o the small number of specimens the mean was calculatedt 410 cycles. However, the small number of specimens (n = 2or 600 N, n = 4 for 500 N) in group EFFS is a limitation of thistudy.

Additionally, it was detected by visual inspection, that withhe exception of two specimens all IPIRFDPs fractured com-letely, which speaks for good adhesion of the veneeringomposite to the framework (examples for typical fracturesee Fig. 5). Besides this fact, it could also be detected that theuccal lamella of the veneering material not only fracturedhen it was stressed eccentrically, but also when stressed

entrically. Dynamic loading, e.g. a leading function duringastication or parafunctions, should be monitored during

linical placement of the restoration.Although in vitro studies are important for preclinical test-

ng, the transferability of the results of an in vitro study tohe clinical situation is limited. It was shown that the modu-us of elasticity of the luting resin and the abutment materialnfluenced the fracture resistance of all-ceramic crowns [33].n accordance with this outcome another study [34] showed,hat fracture force was found to be significantly greater forll-ceramic single crowns fixed on CoCrMo alloy materials1838 N) than for crowns on human teeth (888 N). However,hese studies are transferable only in part to the present study,ecause the support of the abutment material by a full crownnd an inlay retainer differs substantially. However, also inhe present study, the stiffness of the metal abutment mightave influenced the result, because of the different modulusf elasticity of CoCr alloy (180–240 GPa), dentin (15–20 GPa) andnamel (50–85 GPa). Further, it has been demonstrated thatbutment mobility is a decisive factor in the evaluation of frac-ure strength, and when a small amount of abutment rotations allowed, failure is more likely [35]. In this context Kappertt al. reported a mean fracture strength of 703 N for three-unitn-Ceram posterior FDPs with physiological abutment mobil-ty in comparison to 2225 N for the group without abutment

obility [36]. Therefore, according to other studies [37,38] inhe present study a 200 �m thick soft silicone layer was used to

enerate a tooth mobility of 50 �m to simulate a physiologicalooth mobility of grade 0 [39].

To the knowledge of the authors, clinical studies deter-ining the success rate of inlay-retained FDPs with Y-TZP

( 2 0 1 0 ) 67–75 73

frameworks are not available. However, a prospective clinicalevaluation was published recently to show the long-term clin-ical behavior of resin-bonded fixed dental prostheses (RBFDPs)with a retentive, minimally invasive preparation design [40].Since 1985, 232 RBFDPs with a retentive preparation designwere placed and 84 RBFDPs could be re-evaluated in 2005/2006.The total of 12 failures was observed and resulted in a survivalrate of 77% after 10 years. However, only four catastrophic fail-ures occurred. After re-bonding or reparation the survival rateincreased to 88%. Interestingly, RBFDPs made of a titaniumalloy exhibited a statistically significantly higher survival ratethan RBFDPs made of a non-precious CoCr alloy in terms ofall failures. As mentioned before CoCr alloys have a modu-lus of elasticity of 180–240 GPa, which is comparable to Y-TZPceramic (210 GPa). In contrast to that titanium based alloyshave an elastic modulus of 75 GPa. A possible explanation forthe significantly lower failure rate might be that inlay-retainedFDPs made of titanium based alloys are balancing the toothmobility to a certain extent in comparison to rigid materialslike Y-TZP or CoCr alloys.

On the other hand, a prospective clinical cohort study [41]determined the success rate of three to five-unit posteriorFDPs with Y-TZP frameworks after 3 years of function. In thisstudy 36 patients with 46 FDPs were available for examina-tion after 36 months. No fractures occurred, rendering a 100%success rate of the Y-TZP frameworks. Seven FDPs had to bereplaced because of biological and technical problems. Thus,the survival rate was 84.8%. Secondary caries was found in10.9% of the FDPs, and chipping of the veneering ceramic wasfound in 13.0%. A similar study [42] showed comparable results(mean observation time 31 months) for 20 three-unit poste-rior FDPs with a survival rate of 100% and minor chipping ofthe veneering ceramic of 25%. Even though the design of theFDPs was different in these in vivo studies, compared to thepresent in vitro study, all studies showed that Y-TZP frame-works demonstrated sufficient stability for replacement ofposterior teeth. Since the veneering of the restorations seemsto be a limiting factor, an easily reparable composite veneer-ing with wear characteristics similar to human enamel mightbe considered as an option for veneering inlay-retained FDPs.However, the prospective clinical testing of the industriallyprefabricated inlay-retained FDPs has to be performed beforethey can be recommended for general clinical use.

5. Conclusions

Under the limitations of the study the following conclusionscan be drawn:

- Considering the maximum bite forces in the molar region,it seems possible to use yttria-stabilized tetragonal zirconiaceramic with a connector size of 4.7 mm2 as a frameworkmaterial.

- The prefabricated all-ceramic inlay-retained FDPs seemsuitable for restoring missing upper and lower first molars.

- It seems advisable to remove any contact points in eccentricstatic or dynamic occlusion.

l s 2

r

74 d e n t a l m a t e r i a

Acknowledgement

The authors are grateful to Mr. Ulrich Wennemann (Gap-less, Umkirch, Germany) for providing the prefabricatedall-ceramic inlay-retained FDPs used in this study.

e f e r e n c e s

[1] Mumghamba EG, Fabian FM. Tooth loss among habitualchewing-stick and plastic toothbrush users in the adultpopulation of Mtwara, rural Tanzania. Int J Dent Hyg2005;3:64–9.

[2] Cretsi P, Hedderich J, Freitag S, Kern M. Movement andtipping of teeth after loss of first molars. J Dent Res 2005;84.Abstr. No. 2904.

[3] Dean MC. A radiographic and histological study of modernhuman lower first permanent molar root growth during thesupraosseous eruptive phase. J Hum Evol 2007;53:635–46.

[4] Igarashi Y, Yamashita S, Kuroiwa A. Changes in interarchdistance and condylar position related to loss of occlusalsupport for partially edentulous patients. A pilot study. Eur JProsthodont Restor Dent 1999;7:107–11.

[5] Balevi B, Shepperd S. The management of an endodonticallyabscessed tooth: patient health state utility, decision-treeand economic analysis. BMC Oral Health 2007;7:17.

[6] Bergenholtz G, Nyman S. Endodontic complicationsfollowing periodontal and prosthetic treatment of patientswith advanced periodontal disease. J Periodontol1984;55:63–8.

[7] Edelhoff D, Sorensen JA. Tooth structure removal associatedwith various preparation designs for posterior teeth. JProsthet Dent 2002;22:241–9.

[8] Edelhoff D, Spiekermann H, Yildirim M. Metal-freeinlay-retained fixed partial dentures. Quintessence Int2001;32:269–81.

[9] Wolfart S, Bohlsen F, Wegner SM, Kern M. A preliminaryprospective evaluation of all-ceramic crown-retained andinlay-retained fixed partial dentures. Int J Prosthodont2005;18:497–505.

[10] Sjölin R, Sundh A, Bergmann M. The Decim system for theproduction of dental restorations. Int J Comput Dent1999;2:197–207.

[11] Tinschert J, Natt G, Mohrbotter N, Spiekermann H, SchulzeKA. Lifetime of alumina- and zirconia ceramics used forcrown and bridge restorations. J Biomed Mater Res B: ApplBiomater 2007;80:317–21.

[12] Wennemann U. A method for producing dentures or anartificial tooth. World Intellectual Property Organization;2003. Pub. No.: WO/2003/068100.

[13] Katz F. Literature review of zirconia in dentistry and fracturestrength of slot-inlay fixed dental prostheses frameworks.Thesis. Freiburg, Germany: Albert-Ludwigs University; 2007.

[14] Hahn P, Kabel A, Wennemann U. Shear bond strength ofveneering composite to zirconium-oxide ceramic. J Dent Res2006;85(Special Issue B). Abstr. No. 30.

[15] Wichnalek N. A new wax dublication technique. Dent Labor1999;47:1799–810.

[16] Wolfart S, Ludwig K, Uphaus A, Kern M. Fracture strength ofall-ceramic posterior inlay-retained fixed partial dentures.

Dent Mater 2007;23:1513–20.

[17] Kern M, Thompson VP. Sandblasting and silica coating of aglass-infiltrated alumina ceramic: volume loss, morphology,and changes in the surface composition. J Prosthet Dent1994;71:453–61.

6 ( 2 0 1 0 ) 67–75

[18] Quaas AC, Yang B, Kern M. Panavia F 2.0 bonding tocontaminated zirconia ceramic after different cleaningprocedures. Dent Mater 2007;23:506–12.

[19] Yang B, Lange-Jansen HC, Scharnberg M, Wolfart S, LudwigK, Adelung R, et al. Influence of saliva contamination onzirconia ceramic bonding. Dent Mater 2008;24:508–13.

[20] Kern M, Wegner SM. Bonding to zirconia ceramic: adhesionmethods and their durability. Dent Mater 1998;14:64–71.

[21] Christel P, Meunier A, Heller M, Torre JP, Peille CN.Mechanical properties and short-term in-vivo evaluation ofyttrium-oxide-partially-stabilized zirconia. J Biomed MaterRes 1989;23:45–61.

[22] DeGee A, Werner A. Wear of composites K-0166A, K0166B,K-0170, Sinfony, and Z250 of project K-0170:14.1282determined in the ACTA wear machine. Project numberP2008.003 Wear; 2008 a.degree@acta.nl.

[23] Sailer I, Feher A, Filser F, Gauckler LJ, Lüthy H, Hämmerle CH.Five-year clinical results of zirconia frameworks for posteriorfixed partial dentures. Int J Prosthodont 2007;20:383–8.

[24] Tezvergil A, Lassila LV, Vallittu PK. Composite-compositerepair bond strength: effect of different adhesion primers. JDent 2003;31:521–5.

[25] Tylka DF, Stewart GP. Comparison of acidulated phosphatefluoride gel and hydrofluoric acid etchants forporcelain-composite repair. J Prosthet Dent 1994;72:121–7.

[26] Gibbs CH, Mahan PE, Lundeen HC, Brehnan K, Walsh EK,Sinkewiz SL, et al. Occlusal forces duringchewing—influences of biting strength and foodconsistency. J Prosthet Dent 1981;46:561–7.

[27] Linderholm H, Wennstrom A. Isometric bite force and itsrelation to general muscle forge and body build. ActaOdontol Scand 1970;28:679–89.

[28] Waltimo A, Kononen M. A novel bite force recorder andmaximal isometric bite force values for healthy youngadults. Scand J Dent Res 1993;101:171–5.

[29] Körber K, Ludwig K. The maximum bite force as a criticalfactor for fixed partial dentures. Dent Labor 1983;31:55–60.

[30] Castellani D, Baccetti T, Giovannoni A, Bernardini UD.Resistance to fracture of metal ceramic and all-ceramiccrowns. Int J Prosthodont 1994;7:149–54.

[31] Geis-Gerstorfer J, Faeßler P. Evaluation of the endurancelimit for fatique cycling of dental ceramics. Dtsch ZahnarztlZ 1999;54:692–4.

[32] Tinschert J, Natt G, Mautsch W, Augthun M, Spiekermann H.Fracture resistance of lithium disilicate-, alumina-, andzirconia-based three-unit fixed partial dentures: a laboratorystudy. Int J Prosthodont 2001;14:231–8.

[33] Scherrer SS, de Rijk WG. The fracture resistance ofall-ceramic crowns on supporting structures with differentelastic moduli. Int J Prosthodont 1993;6:462–7.

[34] Rosentritt M, Plein T, Kolbeck C, Behr M, Handel G. In vitrofracture force and marginal adaptation of ceramic crownsfixed on natural and artificial teeth. Int J Prosthodont2000;13:387–91.

[35] Kelly JR, Tesk JA, Sorensen JA. Failure of all-ceramic fixedpartial dentures in vitro and in vivo: analysis and modeling.J Dent Res 1995;74:1253–8.

[36] Kappert HF, Knode H, Schultheiss R. Strength of theIn-ceram system under mechanical loads and thermocyclingin artificial saliva. Dtsch Zahnärztl Z 1991;46:129–31.

[37] Kern M, Douglas WH, Fechtig T, Strub JR, DeLong R. Fracturestrength of all-porcelain, resin-bonded bridges after testingin an artificial oral environment. J Dent 1993;21:117–21.

[38] Koutayas SO, Kakaboura A, Hussein A, Strub JR. Colorimetricevaluation of the influence of five different restorativematerials on the color of veneered densely sinteredalumina. J Esthet Restor Dent 2003;15:353–60.

s 2 6

d e n t a l m a t e r i a l

[39] Nyman S, Lindhe J. Examinations of patients withperiodontal desease. In: Lindhe J, Karring T, Lang NP, editors.Clinical periodontology and implant dentistry. 4th ed.

Oxford: Blackwell Munksgaard; 2003. p. 409.

[40] Aggstaller H, Beuer F, Edelhoff D, Rammelsberg P, Gernet W.Long-term clinical performance of resin-bonded fixedpartial dentures with retentive preparation geometry inanterior and posterior areas. J Adhes Dent 2008;10:301–6.

( 2 0 1 0 ) 67–75 75

[41] Sailer I, Feher A, Filser F, Luthy H, Gauckler LJ, Schärer P, et al.Prospective clinical study of zirconia posterior fixed partialdentures: 3-year follow-up. Quintessence Int 2006;37:685–93.

[42] Raigrodski AJ, Chiche GJ, Potiket N, Hochstedler JL,Mohamed SE, Billiot S, et al. The efficacy of posteriorthree-unit zirconium-oxide-based ceramic fixed partialdental prostheses: a prospective clinical pilot study. JProsthet Dent 2006;96:237–44.