An

diff

reli

Matthew R. And

Supported by a Stanley D. Tylman RFirst place, 2012 Tylman Award Pro

aGraduate student, Graduate ProstSeattle, Wash.bProfessor, Department of RestorativcResearch Associate Professor, MatedProfessor, Graduate Prosthodontics

Anderson et al

in vitro study of the effect of

erent restorative materials on the

ability of a veneering porcelain

erson, DMD, MSD,a

Kwok-Hung Chung, DDS, MS, PhD,b Brian D. Flinn, MS, PhD,c

and Ariel J. Raigrodski, DMD, MSd

University of Washington, Seattle, Wash

Statement of problem. Implant-supported, porcelain veneered restorations experience a greater rate of porcelain fracturethan tooth-supported restorations. For completely edentulous patients, one approach to minimizing porcelain fracture isto use acrylic resin in the mandible, although its efficacy is unknown.

Purpose. The purpose of this study was to evaluate the reliability of a veneering porcelain fatigued with different restorativematerials in vitro.

Material and methods. Fifty-nine veneering porcelain disk specimens were fabricated by layering veneering porcelain onnickel-chromium base metal alloy disks. Four groups of different indenter materials fatigued the porcelain specimens: groupWC, tungsten carbide served as a control; group FC, pressed leucite glass ceramic; group NHC, nanohybrid composite resindenture tooth; and group AR, unfilled acrylic resin denture tooth. Porcelain specimens were randomly divided into 4 groups(n¼14). A step-stress accelerated life-testing model was used. Use-level probability Weibull plots were generated, and thereliability of each group was estimated for a theoretical completion of 50 000 cycles at 150 N.

Results. Nanohybrid composite resin and unfilled acrylic resin denture tooth groups had higher reliability than tungstencarbide and leucite glass ceramic groups. No significant differences existed between the reliability of the tungsten carbide andleucite glass ceramic groups and the nanohybrid composite resin and acrylic resin denture tooth groups.

Conclusions. Veneering porcelain disk specimens fatigued with the unfilled acrylic resin and nanohybrid composite resindenture tooth indenters exhibited higher reliability than the specimens fatigued with either the tungsten carbide or leuciteglass ceramic indenters. All of the veneering porcelain disk specimens failed with the same mode of fracture, althoughthe surface posttest exhibited different fracture characteristics among specimens fatigued with the 4 different materials.(J Prosthet Dent 2013;110:521-528)

Clinical Implications

In completely edentulous patients with dual-arch implant-supportedfixed dental prostheses, a different restorative material, such as acrylicresin denture teeth, in one arch should improve the fatigue resistance ofa veneering porcelain in the opposing arch.

Implant-supported fixed dental pros-theses (FDP) are a popular treatmentalternative for completely edentulouspatients.1,2 These fixed restorations are

esearch Ggram, Am

hodontics

e Dentistrrials Scien, Departm

frequently fabricated with either a metalor ceramic infrastructure. Similar tocomplete mouth rehabilitations withtooth-supported FDPs, one approach to

rant from the American Academy of Fixed Proerican Academy of Fixed Prosthodontics.

, Department of Restorative Dentistry, Scho

y, University of Washington School of Dentistrce and Engineering, University of Washington Sent of Restorative Dentistry, University of Was

treating these patients is to veneer theframework of both arches with porce-lain,1,3 even though esthetic dental por-celain is susceptible to premature failure

sthodontics.

ol of Dentistry, University of Washington,

y, Seattle, Wash.chool of Engineering, Seattle, Wash.hington School of Dentistry, Seattle, Wash.

522 Volume 110 Issue 6

with repeated loading in moist condi-tions.4 Moreover, recent studies havedemonstrated a significant increase in theincidence of porcelain fracture inimplant-supported, porcelain veneeredFDPs compared with porcelain veneeredtooth-supported FDPs.5-8

The increased incidence of ve-neering porcelain fracture can beattributed to several factors, includingthe occlusal scheme used in the resto-rations, occlusal interferences, and theparafunctional habits of a patient.5 Inaddition, the lack of a periodontal lig-ament may be a significant contrib-uting factor in the higher incidence ofveneering porcelain fracture.9,10 Theperiodontal ligament provides proprio-ception and helps transmit sensoryfeedback about jaw position duringfunction; its absence has been demon-strated to result in increased mandib-ular muscle discoordination anddecreased tactile sensitivity in edentu-lous patients.11-13 It also permits adegree of tooth movement, from25 mm to 150 mm in periodontallyhealthy teeth and dampens forces thatact on teeth and tooth-supported res-torations compared with implants thatlack mobility.14,15 By comparison,osseointegrated implants typically onlymove up to 10 mm.16,17 Owing tomandibular discoordination, off-axisloading and excursive contacts are tobe expected at a higher rate, and, withthe absence of a force-absorbing func-tion with osseointegrated implants,these forces combined may influencethe rate of veneering porcelain fracturefor implant-supported crowns versussimilar tooth-supported restorations.9

Kinsel and Lin18 conducted a retro-spective analysis of implant-supportedmetal ceramic (ISMC) crowns opposingISMC, tooth-supported metal ceramic,and natural teeth.18 Both patient-specificfactors and implant-specific factorsinfluenced porcelain fractures in thisstudy.When patient-specific factors werecontrolled, the researchers concludedthat an ISMC restoration opposinganother ISMC restoration increased theodds of porcelain fracture by 7 comparedwith an ISMC restoration opposing either

The Journal of Prosthetic Dentis

a tooth-supported metal ceramic resto-ration or natural teeth and that 79% ofmajor porcelain fractureswithin the studyoccurred in the ISMC-ISMC group.18 Therate and types of complications associ-ated with veneered porcelain may beamplified in complete mouth implant-supported rehabilitations.

Intraoral repair of porcelain frac-tures have only limited success, andproblems of color matching and thebonding of repair material maypersist.19-21 An ongoing need for por-celain repair requires frequent mainte-nance and additional cost. Differentmaterials or a combination of them,such as veneering porcelain opposing amore resilient material, may addressthis clinical challenge with ISMC/ISMCrehabilitations.18

The use of acrylic resin denture teethwith implant-supported restorations wasoriginally advocated by Brånemark22 tomitigate occlusal force transmission toimplant parts and the bone-implantinterface.23-25 Although acrylic resinshave been shown to reduce force trans-mission in vitro when compared withdental porcelain andgold,26 these studiesdid not examine the role of a material inthe longevity of a prosthesis. The type ofrestorative material has not been shownto influence implant survival.27

To minimize porcelain fractures,one treatment option is to veneer themaxillary arch infrastructure with por-celain and the mandibular archwith acrylic resin denture teeth.28

The complications associated with us-ing acrylic resin in fixed prosthodontics,for example wear, are well docu-mented.29-32 However, if its use inconjunction with porcelain leads to asubstantial minimization of porcelainfracture, then maintaining one arch asan implant-supported metal acrylicresin prosthesis may be advantageous.

Porcelain fatigue studies commonlyuse indenters such as stainless steel ortungsten carbide (WC) with a highelastic modulus.33,34 Applying resultsfrom such studies to clinical situationsin which porcelain may be interactingwith antagonists of different materialsis difficult. In addition to the size of the

try

contact area, the indenter material canaffect the apparent strength of brittlematerials.35,36 Steel-ball indenters asopposed to glass-ball indenters havebeen shown to lead to a statisticallysignificant increase in fracture strengthin glass slabs. This increase is attributedto the differences in interfacial friction,stress distribution, and elastic con-stants of the 2 materials.37

The purpose of this study was toevaluate the reliability of a veneeringporcelain on metal with different restor-ative materials in an in vitro environmentthat simulated the rigidity of osseointe-grated implants. The first null hypothesiswas that the reliability of the veneeringporcelain would not be affected whenfatiguedbydifferent restorativematerials.The second null hypothesis was that nodifferences would be observed in themode of failure or in the surface andfracture characteristics of the veneeringporcelain after testing.

MATERIAL AND METHODS

Veneering porcelain disk specimenpreparation

Fifty-nine metal disk-shaped speci-mens (10�1.5 mm) were fabricatedwith autopolymerized acrylic resin(Pattern Resin LS; GC America Inc) byplacing the acrylic resin into a custom,milled aluminum mold. After 30 mi-nutes, the disks were removed from themold and invested with a carbon-freephosphate investment (CeraFina;Whip Mix Corp). After the investmenthad set for a minimum of 90 minutes,it was placed in a burnout furnaceat 850�C. A nickel-chromium basemetal alloy (Rexillium V; Jelenko) wasused to cast the metal disks with anonvacuum centrifugal casting machine(Centrifico; Kerr Dental LaboratoryProducts). After devesting and sprueremoval, the disks were airborne-particle abraded with 50-mmaluminum oxide particles (Kavas) for10 seconds at 0.2 MPa pressure and atan approximate distance of 10 mm.The disks were cleaned ultrasonically

Anderson et al

400

300

200

100

020 000 40 000 60 000 80 000

Light

~120 000

ModerateAggressive

Cycles

New

tons

1 Step-stress profiles.

December 2013 523

for 30 minutes in distilled water, steamcleaned, dried, and degassed at 980�Cin a porcelain furnace (Austromat3001; DEKEMA Service GmbH). Aleucite glass veneering porcelain (EX3;Noritake) was hand layered onto themetal disks with a custom-milledaluminum mold. Gauze and light fingerpressure was used to condense thepowder. The porcelain was applied in2 steps. After the first layer wascondensed and fired in a porcelainfurnace at 920�C under vacuum (Aus-tromat 3001; DEKEMA Service GmbH),the specimens were allowed to cool toroom temperature. A second layer ofporcelain was applied by following thesame procedure. Each specimen wasground by using sintered diamond ro-tary instruments (Fine grit laboratorydiamond rotary cutting instruments;Brasseler USA) to a porcelain thicknessof 1.5 mm, followed by hand polishingwith successive polishing instruments(Dialite; Brasseler USA) and a singleglaze firing cycle at 920�C, although noglaze material was applied.

Indenter material preparation

Four different materials served asindenters: (1) group WC, 6.35-mm-diameter WC indenters served as acontrol; this diameter provided a uni-form 0.2 mm2 contact area (r¼0.25mm) as verified with articulating paper;(2) group FC, pressed leucite glassceramic (FC) (IPS InLine PoM; Ivoclar-Vivadent AG); (3) group NHC, nano-hybrid composite resin denture teeth(NHC) (SR Phonares Lingual NHC;Ivoclar-Vivadent AG), which containedinorganic filler; and (4) group AR: un-filled acrylic resin denture teeth (AR)(33� Classic; Dentsply Intl). Each groupcontained 14 antagonist specimens,except for group WC, which containedonly 4 because the material propertiespermitted multiple uses. A light-bodypolyvinyl siloxane (Imprint 3; 3MESPE) impression was made of a WCindenter. Autopolymerizing acrylic resin(Pattern Resin LS; GC America Inc.)was applied into the polyvinyl siloxanemold to a depth of 1.5 mm. With the

Anderson et al

same protocol and materials for metaldisk fabrication, metal posts thatmeasured 3�5 mm were made. Oncepolymerized, the acrylic resin patternwas removed and attached with wax tothe top of a metal post. Wax sprueswere attached to the sides of the resinpatterns. Care was taken not to alterthe tip of the resin patterns. Once thesprues were attached, the patterns wereinvested with phosphate-bonded in-vestment (CeraFina; Whip Mix Corp)and allowed to set for 90 minutes.Once set, the investments were placedin an oven for 40 minutes, and theacrylic resin was eliminated at 850�C.FC (IPS InLine PoM; Ivoclar-VivadentAG) was pressed into the mold ac-cording to manufacturer’s instructions.Once cooled, the FC specimens weredevested and polished with the sameprotocol as the veneering porcelain diskspecimens.

For the denture teeth groups, thepalatal cusps of the maxillary secondpremolars of the NHC and AR dentureteeth were sectioned and placed intospecimen holders. The bases contactedmetal, and autopolymerizing acrylicresin (Pattern Resin LS; GC America)was added between the holder and thesectioned cusps. The FC indenters wereassembled in the same manner. Eachspecimen’s contact area was adjustedby using a diamond rotary cutting in-strument (Brasseler USA) to an area of0.2 mm2 and was measured with digitalcalipers (Digital precision calipers

CP8806-T; Carrera Precision). Only theborders of the contact area wereadjusted to preserve the as-receivedsurface from the manufacturers.

A step-stress accelerated life-testingmodel was used for this study, whichpermitted a product’s reliability to beestimated from a relatively small sam-ple size with reduced testing time.38,39

Data were gathered across different fa-tigue profiles, light, moderate, andaggressive, which were used in a ratio of4:2:1 (Fig. 1).38 This ratio has beenshown to result in data that correspondwell to a material’s actual service life.40

The minimum number of specimens pergroup (n¼14) was determined bydoubling the ratio.41,42 Specimens werenumbered by using software (ResearchRandomizer v4.0; Geoffrey C Urbaniakand Scott Plous)43 and were randomlydivided into 4 groups that corre-sponded to the 4 categories of antag-onist materials. With the exception ofthe WC group, 1 indenter was usedagainst 1 veneering porcelain diskspecimen. To determine the baselineload for testing, 3 porcelain specimenswere subjected to a single load to fail-ure test with the same WC indentersused in the study. The 3 values wereaveraged, and 30% of the mean load tofailure was used as the baseline to begintesting.38 The maximum load appliedwas 400 N, which was the limit of thetesting apparatus.

Each veneering porcelain disk spec-imen was subjected to sinusoidal, cyclic

Table I. Elastic modulus value (E) for materials tested

MaterialTungstenCarbide

PressedLeucite Glass

Ceramic

NanohybridComposite ResinDenture Tooth

Unfilled AcrylicResin Denture

Tooth

Elasticmodulusvalue

530 GPa 67 GPa 2.5 GPa 1 GPa

100 1000 10 000 100 000 1000 000

WCFCNHCAR

Cycles

Unr

elia

bilit

y

99

10

11E+07

2 Use-level probability Weibull plots at 150 N with 2-sided90% confidence bounds. Veneering porcelain disk specimensfatigued with denture tooth indenters, nanohybrid compositeresin denture teeth (NHC), and unfilled acrylic resin dentureteeth (AR) exhibited significantly higher reliability than thosefatigued with either tungsten carbide (WC) or pressed leuciteglass ceramic (FC) indenters. No statistical significanceexisted between WC and FC groups and NHC and ARgroups.

524 Volume 110 Issue 6

loading at 1.5 Hz in a 37�C water bathin a coil cycler electromechanical fa-tigue machine (Fatigue Tester; Proto-Tech) until failure. The loading cycleconsisted of the indenter making con-tact with the specimen, lifting 0.25 mmoff its surface, and then contactingthe surface again at the desired load.The maximum load for all profiles was400 N. The specimens were visuallyinspected every 5000 cycles with alight stereomicroscope (SMZ 1500;Nikon Corp) with �112.5 magnifica-tion. Failure was defined as the pres-ence of radial cracks with either inneror outer cone cracks. Photographsrepresentative of the posttest weremade of the surfaces of porcelainspecimens and the contacting areasof the indenters.

The load and the number of cycles atwhich failures occurred were recorded,and the data were used to generate use-level probability Weibull plots with 2-sided 90% confidence bounds (CB).39

The reliability of each group was esti-mated for a theoretical completion of 50000 cycles at 150 N (Alta Pro 8; Relia-soft). The modulus of elasticity for the 2denture tooth materials was measuredin compression on rectangular speci-mens (2�2�4.5 mm) machined fromthe denture teeth with a universal testingmachine (Instron 5500R; Instron) at acrosshead speed of 0.01 mm/min. Thevalues of the elastic modulus of the WCand FC materials were determined fromprevious studies.44,45

RESULTS

The single load to failure mean(standard deviation) was 222 �32 N.These data were used to develop step-stress profiles with a load range from75 N to 400 N and up to 120 000 cy-cles. All the specimens were cyclicloaded and experienced fracture. Theelastic modulus (E) values are listed inTable I.

The use-level probability Weibullplot was generated at a stress of 150 Nwith 90% CB (Fig. 2) and revealed nosignificant differences among thegroups because the CB overlapped at

The Journal of Prosthetic Dentis

lower stresses and cycles. However, athigher stress loads and cycles, the CBof groups WC and FC did not overlapwith groups NHC and AR, whichindicated statistically significant dif-ferences among the groups. No signif-icant differences existed in reliabilitybetween the paired groups WC and FCand the groups NHC and AR. Theslope of the Weibull plot, called theshape parameter, is denoted by avalue, b. This value describes the fail-ure rate of the specimens over time.Values <1 signify early failures that aretypically the result of absolute flaws,values of approximately 1 denote fail-ure rates that do not vary with timeand are random in nature, and values

try

>1 indicate failure rates that are theresult of cumulative stress and dam-age.46 All 4 groups had b values >1,which indicated that the damage tothe porcelain disks was cumulative(Table II).

The calculated reliability for atheoretical mission of 50 000 cycles at150 N with 2-sided 90% CB of theporcelain specimens fatigued with thedifferent indenters is listed in Table II.All of the WC and nearly all of the FCfatigued specimens showed failures,whereas approximately 92% of theNHC group and nearly 81% of the ARgroup were free of complications orfailure at the end of a theoretical loadof 150 N for 50 000 cycles.

Anderson et al

Table II. Results of calculated 90% confidence bounded intervals for 50 000cycles at 150 N for tested groups

MaterialTungstenCarbide

PressedLeucite Glass

Ceramic

NanohybridComposite

Resin DentureTooth

Unfilled AcrylicResin Denture

Tooth

Reliability upperconfidencebound

0.253 0.007 0.989 0.951

Reliability lowerconfidencebound

1.065 E-10 7.004 E-26 0.626 0.626

Reliability 0.001 0.004 0.813 0.813

b value 2.58 3.98 2.63 1.66

December 2013 525

Representative veneering porcelaindisk specimens were examined after fa-tigue testing and revealed failure by innerand outer cone cracks for all groups(Fig. 3). Differences were noted in the

3 Posttest photographs of representativespecimen, exhibiting inner and outer connounced circular hackle zone. C, Nanohybor flaking, whereas lower arrow points to

Anderson et al

fracture characteristics among speci-mens fatigued with the 4 different ma-terials. Pronounced circular zones ofhackle were noted in the porcelainspecimens fatigued with FC indenters

porcelain disk specimens at �112.5 mage cracks. B, Pressed leucite glass ceramic frid composite resin denture teeth fatiguedradial crack. D, Unfilled acrylic resin dent

(Fig. 3B). Those specimens fatigued withNHC indenters had a greater amount ofsurface spar, or flakes, than the othergroups and exhibited surface radialcracks that formed perpendicular to thecontact area (Fig. 3C). The indenterposttest contacting surfaces were alsoinspected, and differences were observedfor each of the 4 groups (Fig. 4). TheWCindenters (Fig. 4A) did not changediscernibly, whereas the FC indenters(Fig. 4B) exhibited slight scratches alongtheir surfaces. The NHC indentersexhibited small chips and minor wearwith cracks that radiate from the con-tacting surface through the body of thematerial (Fig. 4C). Within this study, theAR indenters exhibited the greatestamount of wear, and their contactingsurfaces were characterized as beingsmooth and covered with multiplesmall indentations with no chipping(Fig. 4D).

nification. A, Tungsten carbide fatiguedatigued specimen; arrow points to pro-specimen; upper arrow points to spar,ure teeth fatigued specimen.

4 Posttest photographs of representative indenters at �112.5 magnification. A, Tungsten carbide. B, Pressed leucite glassceramic. C, Nanohybrid composite resin denture teeth; chip is evident in upper left, and cracks are radiating from contactsurface. D, Unfilled acrylic resin denture teeth; small indentations are evident across smooth surface.

526 Volume 110 Issue 6

DISCUSSION

This study evaluated the effect ofdifferent indenter materials on the axialloading fatigue of layered veneeringporcelain on metal disks in a wet envi-ronment. The different indenter mate-rials led to significant differences in thereliability of the veneering porcelain diskspecimens between those fatigued withthe WC and FC groups and with theNHC and AR groups. Therefore, thefirst null hypothesis was rejected. Allspecimens failed because of inner andouter cone cracks, although differentfracture characteristics and surfaceappearance also were noted betweenthese 2 sets of groups. Therefore, thesecond null hypothesis was partiallyrejected.

Any changes to the indenters them-selves after cyclic loading were notquantified in this study. However,posttest visual inspection of the 4

The Journal of Prosthetic Dentis

types of indenters did reveal physicalchanges. No material loss, cracks, orfractures were seen in the WC and FCindenters. The NHC indenters exhibitedmild wear on the contacting surfaceand cracks that originated from thecontact point and radiated out throughthe body of the material were observedafter the test. None of the NHC in-denters experienced a chip or grossloss of material during the test. TheAR indenters did not exhibit anycracks, although this group exhibitednoticeable loss of material. The AR in-denters that exhibited the largest vol-ume loss of material after testing hadcontact areas approximately 3 timestheir original size.

The process of fretting, which is aspecial wear process in which 2 bodies inrepeated load are subjected to minutemovements and vibrations and whichleads to material loss, may account forthe increase in contact areas in both

try

the NHC and AR indenters. The axialloading of the specimens did not have ahorizontal, or sliding, component,although it is reasonable to assume thata degree of slight movement of the pis-ton occurred during loading. Theindenter material loss witnessed wasconsistent with previous studies thatdocumented the wear of NHC and ARagainst porcelain surfaces.31,32

Within this study, the veneering por-celain disk specimens fatigued with the2 denture tooth indenters exhibited thehighest reliability. Noticeable differencesexisted between both the fracture char-acteristics of the porcelain disk speci-mens fatigued with the 2 kinds ofindenters and the indenters themselves.The amount of flaking seen on the sur-face of the NHC-fatigued specimens andthe presence of cracks running throughthe body of the NHC indenters may beof concern. If these characteristic find-ings are an indication of how these 2

Anderson et al

December 2013 527

materials affect one another when ven-eered on implant-supported FDPs in anin vivo scenario, then the use of NHCmaterials may not decrease the inci-dence of porcelain-associated compli-cations. The inorganic fillers typicallyused in NHCmaterials may decrease thefracture resistance of porcelain. Forthese reasons, using an unfilled acrylicresin may bemore advantageous despitethe absence of statistical significance inreliability between the 2 denture toothgroups and despite both having similarelastic moduli.

The limitations of this study includethemethodanddesignof theprotocol forfatiguing the veneering porcelain diskspecimens. Vertical loading perpendic-ular to a flat test surface inwhich contact,load, and lift-off are cyclically performedapproximates intraoral contacts in vitro,although it does not simulate the range offorces that can be generated by a patientduring function and parafunction. Inaddition, flat specimens do not replicatethe complex geometry of anatomicallycontoured restorations. Another limita-tion is the contact area. Although it islikely that contact area surfaces would beadjusted in vivo, theNHCandARdentureteeth with the as-purchased surfacetexture were tested. Only the periphery ofthe contact areawasmodified. Therefore,uniform contact geometry was not stan-dardized across all groups. Although thisapproach may have influenced somefindings, it was done to avoid the inclu-sion of additional design variables thatmay result from altering the contactingsurface texture and consistency.

Presently, few studies have examinedthe complication rates of implant-supported FDPs in edentulous pa-tients.8,29,30 The authors are not awareof a study that investigates the prostho-dontic complications of metal ceramicopposing metal acrylic resin implant-supported FDPs in vivo. As moreedentulous patients seek implants tofacilitate their oral rehabilitation, clinicalstudies that examine the influence ofvarious restorative materials on compli-cations and outcomes are indicated. Inthe interim, research on this topic shouldencompass the influence of specific

Anderson et al

adjustment and the polishing protocolsof denture teeth on the fracture resis-tance of porcelain. Combinations ofdifferent ceramics and other commonrestorative materials such as pressableglass ceramics and highly cross-linkedacrylic resins and other popular restor-ative materials such as those found inceramic restorations are also indicated.Currently, this study indicates that usinga different restorative material in onearch to improve the fatigue resistance ofthe veneered porcelain in another mayhave merit. If such an approach is taken,then a material with greater wear prop-erties is more desirable than materialsthat are either brittle or have greater wearresistance.

CONCLUSIONS

Within the limits of this in vitrostudy, the following conclusions may bedrawn:

1. Veneering porcelain disk speci-mens fatigued with the AR and NHCdenture tooth indenters exhibited higherreliability than the specimens fatiguedwith either the WC or FC indenters.

2. All of the veneering porcelaindisk specimens failedwith the samemodeof fracture, although theposttest surfacesexhibited different fracture characteristicsamong specimens fatigued with the 4different materials.

REFERENCES

1. Schwartz-Arad D, Chaushu G. Full-archrestoration of the jaw with fixed ceramometalprosthesis. Int J Oral Maxillofac Implants1998;13:819-25.

2. Chaushu G, Schwartz-Arad D. Full-archrestoration of the jaw with fixed ceramo-metal prosthesis: late implant placement.J Periodontol 1999;70:90-4.

3. Mitrani R, Phillips K, Escudero F. A simplifiedapproach in the fabrication of an implant-supported, full mouth, fixed metal-ceramicrestoration. Pract Proced Aesthet Dent2004;16:125-7.

4. Lawn BR. Fracture of brittle solids. 2nd ed.New York: Cambridge University Press; 1993.p. 180-91.

5. Linkevicius T, Vladimirovas E, Grybauskas S,Puisys A, RutkunasV. Veneer fracture in implant-supported metal-ceramic restorations: part1doverall success rate and impact of occlusalguidance. Stomatologia 2008;10:133-9.

6. Pjetursson BE, Tan K, Lang NP, Brägger U,Egger M, Zwahlen M. A systematic review ofthe survival and complication rates of fixedpartial dentures (FPDs) after an observationperiod of at least 5 years. Clin Oral ImplantsRes 2004;15:625-42.

7. Pjetursson BE, Brägger U, Lang N,Zwahlen M. Comparison of survival andcomplication rates of tooth-supported fixeddental prostheses (FDPs) and implant-supported FDPs and single crowns (SCs).Clin Oral Implants Res 2007;18:97-113.

8. Bozini T, Petridis H, Tzanas K, Garefis P.A meta-analysis of prosthodontics complica-tion rates of implant-supported fixed dentalprostheses in edentulous patients after anobservation period of at least 5 years. Int J OralMaxillofac Implants 2011;26:304-18.

9. Brägger U, Aeschlimann S, Bürgin W,Hämmerle CH, Lang NP. Biological andtechnical complications and failures withfixed partial dentures (FPD) on implants andteeth after four to five years of function. ClinOral Implants Res 2001;12:26-34.

10. El-Sheikh AM, Hobkirk JA, Howell PG,Gilthorpe MS. Passive sensibility in edentu-lous subjects treated with dental implants: apilot study. J Prosthet Dent 2004;91:26-32.

11. Guichet NF. Biological laws governing func-tions of muscles that move the mandible:part I. Occlusal programming. J ProsthetDent 1977;37:648-56.

12. Hämmerle CH, Brägger U, Lussi A,Karayiannis A, Joss A, Lang NP. Threshold oftactile sensitivity perceived with dentalendosseous implants and natural teeth. ClinOral Implant Res 1995;6:83-90.

13. TartagliaGM, Testori T, PallaveraA,Marelli B,Sforza C. Electromyographic analysis ofmasticatory and neck muscles in subjects withnatural dentition, teeth-supported andimplant-supported prostheses. Clin Oral Im-plants Res 2008;19:1081-8.

14. Daly CH, Nicholls JI, Kydd WL, Nansen PD.The response of the human periodontal lig-ament to torsional loading e I. Experimentalmethods. J Biomech 1974;5:517-22.

15. Picton DC, Wills DJ. Viscoelastic propertiesof the periodontal ligament and mucousmembrane. J Prosthet Dent 1978;40:263-72.

16. Sekine H, Komiyama Y, Hutta H, Yoshida K.Mobility characteristics and tactile sensitivity ofosseointegrated mobility characteristics andtactile sensitivity of osseointegrated fixture-supported systems. In: van Steenberghe D, ed-itor. Tissue-integration inoral andmaxillofacialreconstruction. Amsterdam: Exerpta Medica;1986. p. 326-32.

17. Nyman SR, Lang NP. Tooth mobility and thebiological rationale for splinting teeth.Periodont 2000;1994;4:15-22.

18. Kinsel RP, Lin D. Retrospective analysis ofporcelain failures of metal ceramic crownsand fixed partial dentures supported by 729implants in 152 patients: patient-specific andimplant specific predictors of ceramic failure.J Prosthet Dent 2009;101:388-94.

19. Walton JN, MacEntee MI. A retrospectivestudy on the maintenance and repair ofimplant-supported prostheses. Int J Prostho-dont 1993;6:451-5.

528 Volume 110 Issue 6

20. Yanikoglu N. The repair methods for frac-tured metal-porcelain restorations: a reviewof the literature. Eur J Prosthodont RestorDent 2004;12:161-5.

21. Kupiec KA, Wuertz KM, Barkmeier WW,Wilwerding TM. Evaluation of porcelain sur-face treatments and agents for composite-to-porcelain repair. J Prosthet Dent 1996;76:119-24.

22. Brånemark PI. Osseointegration and itsexperimental background. J Prosthet Dent1983;50:399-410.

23. Davis DM, Rimrott R, Zarb GA. Studies onframeworks for osseointegrated prostheses:part 2. The effect of adding acrylic resin orporcelain to form the occlusal superstructure.Int J Oral Maxillofac Implants 1988;3:275-80.

24. Cibirka RM, Razzoog ME, Lang BR,Stohler CS. Determining the force absorptionquotient for restorative materials used inimplant occlusal surfaces. J Prosthet Dent1992;67:361-4.

25. Hobkirk JA, Psarros KJ. The influence ofocclusal surface material on peak mastica-tory forces using osseointegrated implant-supported prostheses. Int J Oral MaxillofacImplants 1998;13:781-90.

26. Gracis SE, Nicholls JI, Chalupnik JD,Yuodelis RA. Shock-absorbingbehavior of five restorative materialsused on implants. Int J Prosthodont 1991;4:282-91.

27. Brunski JB, Puleo DA, Nanci A. Biomaterialsand biomechanics of oral and maxillofacialimplants: current status and future de-velopments. Int J Oral Maxillofac Implants2000;15:15-46.

28. Balshi TJ, Wolfinger GJ. Teeth in a day for themaxilla and mandible: case report. ClinImplant Relat Dent Res 2003;5:11-6.

29. Purcell BA, McGlumphy EA, Holloway JA,Beck FM. Prosthetic complications inmandibular metal resin implant-fixedcomplete dental prosthesis: a 5 to 9 yearanalysis. Int J Oral Maxillofac Implants2008;23:847-57.

The Journal of Prosthetic Dentis

30. Hemmings KW, Schmitt A, Zarb GA. Com-plications and maintenance requirements forfixed prostheses and overdentures in theedentulous mandible: a 5-year report. Int JOral Maxillofac Implants 1994;9:191-6.

31. Ekfeldt A, Oilo G. Occlusal contact wear ofprosthetic materials. An in vivo study. ActaOdontol Scand 1988;46:159-69.

32. Ghazal M, Yang B, Ludwig K, Kern M. Two-body wear of resin and ceramic denture teethin comparison to human enamel. Dent Mater2008;24:502-7.

33. Lawn BR, Deng Y, Thompson VP. Use of con-tact testing in the characterization and designof all-ceramic crownlike layer structures: a re-view. J Prosthet Dent 2001;86:495-510.

34. Kim B, Zhang Y, Pines M, Thompson VP.Fracture of porcelain-veneered structures infatigue. J Dent Res 2007;86:142-6.

35. Baran GR, McCool JI, Boberick KG,Zhang HQ. Size effect in resin / glass com-posite flexural strengths. J Oral Rehabil1999;26:775-80.

36. Green DJ. An introduction to the mechanicalproperties of ceramics. Cambridge: Cam-bridge University Press; 1998. p. 276-9.

37. Johnson KL, O’Conner JJ, Woodward AC.The effect of the indenter elasticity on theHertzian fracture of brittle materials. Proc RSoc Lond A 1973;334:95-117.

38. Nelson W. Accelerated testing: statisticalmodels, test plans, and data analysis. NewYork: Wiley; 1990. p. 493-520.

39. Zhao W, Elsayed EA. A general acceleratedlife model for step-stress testing. IEEE TransReliabil 2005;37:1059-69.

40. Meeker WQ, Hahn GJ. Asymptotically opti-mum over-stress tests to estimate the survivalprobability at a condition with a low expectedfailure probability. Technometrics 1977;19:381-99.

41. Silva NR, de Souza GM, Coelho PG,Stappert CF, Clark EA, Rekow ED, et al. Ef-fect of water storage time and compositecement thickness on fatigue of a glass-ceramic trilayer system. J Biomed Mater Res BAppl Biomater 2008;84:117-23.

try

42. Coelho PG, Silva NR, Bonfante EA,Guess PC, Rekow ED, Thompson VP. Fa-tigue testing of two porcelain-zirconia all-ceramic crown systems. Dent Mater2009;25:1122-7.

43. Urbaniak GC, Plous, S. Research Randomizer(v4.0). Available at: http://www.randomizer.org/. Accessed March 8, 2012.

44. Liu K, Li XP, Rahman M, Liu XD. CBN toolwear in ductile cutting of tungsten carbide.Wear 2003;225:1344-51.

45. Gonzaga C, Cesar P, Okada CY, Fredericci C,Neto FB, Yoshimura HN. Mechanical prop-erties and porosity of dental glass-ceramichot-pressed at different temperatures. MaterRes 2008;11:301-6.

46. Abernethy R. The new Weibull handbook.5th ed. North Palm Beach: Robert B.Abernethy; 2006. p. 2-10.

Corresponding author:Dr Kwok-Hung ChungDepartment of Restorative DentistryUniversity of Washington1959 NE Pacific Street, D-770 HSCBox 357456Seattle, WA 98195-7456.E-mail: akchung@u.washington.edu

AcknowledgmentsThe authors thank Karen Verina for assistance inobtaining photographs of representative speci-mens and for measuring the modulus of elasticityof the test materials and Lloyd Mancl, PhD, forstatistical consultation.

Copyright ª 2013 by the Editorial Council forThe Journal of Prosthetic Dentistry.

Anderson et al