operative Dentistry

The effect of cuspal coverage on the fracture resistance of teethrestored with indirect composite resin restorationsF. J. T. Burke* / N. H. F. Wilson* / D. C. Watts*

The effeet of cu.spal coverage on the in vitro fracture resistance of teeth restored withindirect composite resin inlays was examined. After standardized preparation differingonly in cavity width, two groups of teeth were restored with indirect composite resinrestorations. Two further groups of teeth were prepared in a similar manner, exceptthat the buccal and palatal cusps were redueed by 2 mm, and cuspal-coverage indirectcomposite resin restorations placed. In another group, indirect composite resin restora-tions with only palatal cu.spal coverage were placed. One group of intact teeth acted asa control. Forces were applied to the restored teeth by means of a 4-mm steel bar in aUniversal testing machine. For each of the preparation designs assessed, the fractureload for onlays was greater than that for equivalent inlays: the onlay fracture strengthivas equivalent, for some preparation designs, to that of sound teeth.{Quintessence Int ¡993:24:875-880.)

Introduction

The development of new restorative systems necessi-tates assessment of the physical properties of the com-ponent materials and evaluation of the behavior andperformance of the systems in clinical service. Along-side such investigations it is necessary to determinethe applications and techniques appropriate for thesystems. For composite resin inlay/onlay systems' (ie,indirect composite resin restorations), a restorativetechnique used by 10% of respondents in a recentsurvey in the USA,' in common with other forms ofindirect tooth-colored restorations, it is considered tobe of particular importance to establish guidelines fortooth preparation. Accordingly, the effects of varia-tions in preparation taper,^ width, and depth"* havebeen examined. However, other features of indirectcomposite resin restoration preparations remain to be

' Department ofRestorative Dentistry, University Dental Hospitalof Manchester, Higher Cambridge Street, Manchester M15 6FH,England.

investigated, notably the effects of cuspal coverage,which for many years has been known to greatly in-crease the fracture resistance of teeth restored withrestorations of cast gold.-̂

The purpose of the present in vitro investigationwas to examine the effect of cuspal coverage on thefracture resistance of maxillary premolar teeth re-stored with secondary-cured indirect composite resinrestorations of a fine-particled hybrid material.

Method and materials

Sixty maxillary premolar teeth, free of restorations,defects, caries, and cracks, were divided at randominto six groups of 10 teeth each. None of the teethincluded were of atypical morphology, and the meanmaximum buccolingual width {ie, the distance fromthe maximal convexity on the buccal surface to themaximal convexity of the palatal surface) of the teethwithin and between each group varied by no morethan 2.5%. Following 24 hours" postextraction stor-age in buffered formal saline, the teeth were stored inroom-tempera ture water (20"C) at all times other thanwhen necessary to dry the teeth for steps of the ex-perimental procedure. Each tooth was fixed, long axis

Quintessence International Volume 24, Number 12/1993 875

Operative Dentistry

Fig 1 Onlay preparations utilized in the investigation, withhorizontai reduction of the cusp by 2 mm.

Code Descr ipt ion

I Minimal fracture of tooth

or inlay

ml II Less than half ofinlay lost

III Inlay fracture through midline;

half of inlay displaced or lost

IV More than half ofiniay lost

V Severe fracture of tooth

and/or inlay

Fig 2 Classification of fracture.

vertical and crown uppermost, in a stainless steel mold(15 X 15 X 15 mm) with a central cylindrical hole12 mm in diameter, by using a light-curing compositeresin restorative that extended to within 2 mm of theeementoenamel junction.

An inlay preparation of standardized design, basedon an examination of "typical" inlay preparations inextracted teeth,'' was adapted for use in this investi-gation. The proximal hox width was one half the huc-colingual width. The occiusal isthmus width (meas-ured midway between the mesial and distal surfacesof the tooth) differed among the groups (see beiow).The occiusal isthmus depth, measured from the huccalcusp tip, was 4 mm. The cavity wall tapered 4 degrees.The gingival margin was located 1 mm above the ee-mentoenamel junction. The gingival floor was 2 mmdeep, and the pulpoaxial line angle was beveled.

Based on this design, preparations were completedin the teeth included in each group:

Group 1.1: Occiusal width of the preparation wasone third the buccolingual width.

Group 1.2: Occiusal width of the preparation wasone haif the buccolingual width.

Group 2,1: Occlusai width of the preparation wasone third the buccohngual width; buccal and palatalcusps were reduced by 2 mm (Fig I).

Group 2,2: Occiusal width of the preparation wasone half the buccolingual width; buccal and palatalcusps were reduced by 2 mm.

Group 2.3: Occiusal width of the preparation wasone half the buccolingual width; only the palatal cuspwas reduced by 2 mm.

The dimensions of the preparations were calculatedfrom the previously measured tooth dimensions. Theinitial phase of each preparation was completed witha parallel-sided, round-ended fissure diamond held ina high-speed handpiece operating with water coolantA 4-degree taper was then added using a Bredenl ta-pered diamond held in a laboratory handpiece oper-ating at 8,000 rpm and fixed in a Bachmann designparallelometer (Cendrés & Métaux). A hase of Ketac-Bond (ESPE) was then placed to cover all exposeddentin, and, once set, was trimmed to a uniform thick-ness of 0.5 mm across the pulpal floor. Conventionalinstruments and techniques were used throughout; theoperator wore x2,5 magnifying loupes.

A one-stage poly(vinyl siloxane) putty and light-hodied paste impression (Prestdent, Coltene) wastaken of each prepared tooth. Casts were poured inVelmix stone (Kerr/Sybron) within 1 to 24 hours ofrecording the impression. The composite resin resto-

8/6 Quintessence International Volume 24, Number 12/1993

operative Dentistry

rations were fabricated by one technician employinga standardized technique. A fine-particled hybrid in-direct composite resin system (Brilliant, Coltene) wasused according to manufacturer's recommendations.

Each restoration was returned from the laboratoryon its die. The marginal fit of the restoration wasexamined on the die and the tooth. Any restorationconsidered unsatisfactory was rejected and remadefollowing a repeat impression. After cleaning andthorough drying of the tooth with oil-free air, theenamel was etched with Coltene etching gel (Cohene)for 60 seconds, washed for 30 seconds under a jet ofwater, and then thoroughly dried. The fitting surfaceof each restoration was cleaned by application of Col-tene etching gel on sponge pellets, followed by a 30-second washing and 30-second drying cycle. The in-lays were luted according to manufacturer's instruc-tions using a dual-curing luting system (Coltene Duo-Cure), In each case, the adhesive was light cured witha light-curing unit (Luxor: ICI Dental) for 40 secondsfrom each of three directions (mesial, distal, and oc-clusal). Excess luting resin was removed with 40-^mgrit diamond finishing burs (Composhape: Intensiv)operating at high speed with water coolant.

Eoliowing storage under water at room temperaturefor a minimum of 4 hours, to allow for any postcuringof the luting agent that might have occurred, the re-stored teeth were subjected to compressive loading. Astainless steel bar 4 mm in diameter was applied cen-trally along the mesiodistal axis of the occlusal surfaceof the restoration at a crosshead speed of 1 mm/miniti a universal testing maehine. The force (in Newtons)required to induce fracture was recorded, as was themode of fracture according to a previously describedclassification* (Fig 2). One group of 10 sound, unpre-pared maxillary premolar teeth, of similar dimensionsto the test groups, was tested, to serve as a control.

Results

Results of the fractui-e testing of the onlay groups andthe equivalent inlay groups are shown in Fig 3. Theseresults indicated that, for each of the preparation de-signs tested, the fracture load for inlays was less thanthat for the equivalent onlays: the value for the onlaypreparations was equivalent to that for sound teethfor the onlay group 2.2. In all eases, the fracture loadsfor onlays were significantly greater [P <.O1) thanthe fracture loads for teeth restored with the equiva-lent inlays.

Fig 3 Fracture strengths of the inlay and onlay prepara-tions restored with indirect composite resin restorations.(BLW) Buccoiingual width.

Comparisons of fracture behavior and fracture loadare shown in Figs 4 and 5.

DiscussioD

Investigations of this nature have been used over aperiod of 30 years to assess the strength of restoredteeth.' Such tooth fracture methods have recentlybeen criticized as lacking discrimination,' althoughwhen standardized teehniques are used and the onlyvariable is cavity design, it may be expected that theresults would indicate the degree of tooth weakeningor strengthetiing associated with each cavity designvariable. In the present investigation, care was takento standardize the simulated clinical and technicalprocedures, despite the considerable time required forproduction of each specimen (circa 3 hours). Accord-ingly, the only variable present other than cavity de-sign was the natural variation in tooth morphology.Steps were taken during selection of teeth to keep suchvariation to a minimum.

The use of a 4-mm-diameter steel bar for fracturetesting was considered to facilitate reproducible load-ing from one specimen to another. A steel sphere wasused in an initial series of tests, but was found to bedifficult to maintain in position on the teeth. Otherinvestigators have hollowed out the occlusal surfacesof restorations to allow placement of such spheres,*"'"but it was considered that such preparation mightalter loading and influence the interpretation of thefindings. Accordingly this option was not considered.It was also considered that the use of a 4-mm spherewould allow the fracturing force to be applied to the

Quintessence Internalionai Voiume 24, Number 12/1993 877

operative Dentistry

1,2 -f

0,6 -

0.4 -

0 , 2 -

/A

^ \

¿71f

III IV

Fracture mode

V Sound teeth

i Onlay D Comparable inlay

Fig 4 Fracture behavior of onlays(buccal and palatal cuspal cover-age) and inlays (one third bucco-lingual width).

1,6-f

/A

¿7IV V Sound teeth

Fracture mode

I Onlay G Comparable inlayFig 5 Fracture behavior of onlays(buceal and palatal cuspal cover-age) and inlays (one half bucco-iingual width).

878 Quintessence International Volume 24, Number 12/1993

operative Dentistry

restoration in all cases, rather than to the tooth, ortooth and restoration.

The resnits of investigations carried out on teethrestored with cast gold mesio-occlusodistal onlays^ in-dicate that teeth restored with such restorations aretwice as strong as unrcstored teeth with the same cav-ity preparation. Salis et al," in 1987. reported that thefracture resistance of teeth restored with mesio-occlu-sodistal gold inlays with cuspal coverage is enhancedto a level much greater than that of sound teeth. Anondestructive method developed by Malcolm'- andMalcolm and Hood" nses strain gauges honded tothe buccal and palatal surfaces of premolars to meas-ure cuspal movement. These workers have demon-strated that gold overlays, irrespective of design, re-duce eusp deformation to less than that of intact teeth.Furthermore, studies using photoelastic analysis havedemonstrated that occlusal coverage by onlays re-duces stresses in the remaining tooth structure.'" Suchtooth strengthening has also been noted when otherrestorative materials have been used, for example,amalgam." The reasons for the enhanced protectionprovided by cuspal coverage with cast gold have beendiscussed by Sahs et al." These workers concludedthat improved fracture resistance is due to the con-fining effect of the overlying cast restoration and tothe behavior of gold when subjected to an impactforce." However, because comparable results havebeen obtained in the present investigation, thestrengthening effect cannot be attributed solely to theuse of gold, and enhanced fracture resistance maytherefore be considered to be due to the confiningeffect of the restoration. Indeed, it may be consideredthat a restoration such as a composite resin onlay,which is bonded to tooth structure, may be particu-larly suitable in this respect.

The effect uf indirect composite resin restorationson cuspal reinforcement has been examined by Lopeset al,'̂ who, using an adaptation of the method uti-lized by Morin et al,'" found that the mean relativestiffness of teeth restored with postcured inlays (P-50,3M Dental) was 0.98 of the mean value for soundteeth. This finding was explained by the high bondstrength achieved between restoration and tooth andby consideration that the po.stcure procedures hadconferred a high degree of stiffness on the restorativematerial.

The cusp fracture resistance of teeth restored withindirect composite resin restorations has also beenexamined by Wendt," using a technique in whichteeth were restored with direct and indirect composite

resin re.storations and various methods of placement.His results indicate that cusp fracture resistance forany restored teeth is significantly less than that ofsound controls; three groups of "heat-treated" inlaysdid not demonstrate significantly different fracture re-sistance from a direct placement group. However, be-cause the cavities used in the work reported byWendt" were narrower than those considered typicalfor indirect composite resin inlays,' and the inlayswere placed with a low-viscosity enamel bonding resinrather than the luting agents provided for use withcomposite resin inlay systems, the techniques usedwere probably not typical of those the conditions em-ployed in clinical practice.

The reduction in fracture resistance of teeth re-stored with composite resin inlays shown in the pres-ent work should be examined in relation to the resultsof the work by Lopes'̂ in which composite resin inlaysrestored the stiffness of teeth without the use of cuspalcoverage. Conversely, euspal coverage was necessaryto restore fracture resistance of the teeth utilized inthe present investigation. Such an apparent contra-diction may be explained by the different experimentalmethods used, in that the ultimate crushing strengthof a tooth is unlikely to be achieved in the oral cav-ity.'̂ The strain measurement method used by Lopes'*would therefore appear to estimate a different para-meter of tooth function than the fracture method uti-lized in the present work and in the study by Wendt."In the work reported by Lopes,'* fatiguing forces wereapplied to the tooth in a manner similar to everydaymasticatory forées that cau.se flexure; the bonded res-toration may prevent such fiexure. In the latter in-vestigations, a single application of force caused frac-ture in a manner similar to unexpected biting on ahard object. Both types of investigation therefore baveclinical relevance, since tooth fractures seen in theclinical situation may occur with or without obviousreason (Burke FJT, unpublished data).

Coverage of only the palatal cusp resulted in animproved fracture resistance compared to the equiv-alent results for teeth with no cuspal coverage. Thenumber of severe mode V fractures was limited to onein group 2.3, while there were six such fractures in thecomparable group with no cuspal protection. It maytherefore be considered that coverage of the at-riskpalatal cusp may provide sufficient protection fromfracture, while it allows a more conservative prepa-ration than that required for full coverage.

When the fracture mode of the teeth with onlays isconsidered, the results indicate that coverage both re-

Quintessence International Volume 24. Number 12/1993 879

Operative Dentistry

duced fracture resistance and may have reduced cat-astrophic mode V fractures. Indeed, only one suchfracture occurred in onlay group 2.1, none occurredin onlay group 2.2. and one occurred in onlay group2.3. There were three mode V failures in the equivalentgroup of inlays (group 1.1) and six in the group ofinlays with the wider preparations (group 1.2). Thegreatest reduction in the number of severely fracturedspecimens therefore occurred in the groups in whiehthe preparations were widest, at one half the buccol-ingual width, indicating that for composite resin in-lays of the type under investigation, cuspal coverageis advisable when the preparation is greater than orequal to half the buccolingual width.

In the present investigation, the onlay preparationswere standardized to a horizontal cuspal reduction of2 mm. Other types of cuspal coverage may achieve asimilar result, although it has been reported that2 mm is the minimal thickness that will provide ad-equate resistance to occlusal forces for the materialexamined.* However, alternative preparations, such asthose with a reverse bevel on the buccal surface, mayalso provide improved resistance and retention. Theplacement of a groove in the dentin exposed by cuspreduction and the incorporation of a pin channel inthe preparation have also been suggested as methodsof countering the substantial shear forces applied toonlay restorations during heavy loading, particularlyduring lateral excursions.' Further research is re-quired to fully evaluate these approaches.

The improvement in fracture resistance provided bythe onlay must, however, be balanced against the in-creased tooth preparation and longer cavosurfacemargin created when cusps are covered. The resultsof the present study appear to indicate, however, thatthe increased preparation is justified for the materialunder test. Furthermore, the removal of the occlusal2 mm of cusp is unlikely to involve more than minimaldentin preparation and is therefore unlikely to presenta potential irritant to the pulpal tissues.

Summary

In this investigation, the effect of cuspal coverage wasevaluated by comparing the fracture resistance ofteeth restored with a composite resin inlay with thatof teeth restored with a diniensionally similar onlayrestoration. In all cases, the fracture loads were sta-tistically greater iP < .01) for the teeth restored withonlays than for teeth restored with equivalent inlays.The fracture strength of the teeth restored with onlays

was found to be similar to that of sound teeth. Forthe indirect composite resin system investigated, thefracture resistance of prepared teeth is therefore re-stored to a greater degree by onlays than by inlays.

Despite the extent of tooth preparation required,the provision of an indirect composite resin onlayrather than an inlay should be considered when thepotential for cuspal fracture is considered to be high.

Acknowledgments

The authors wish 10 ihank Coltcne AG, Altstatten, Switzerland, fortheir financial support of this project.

References

1. Burke FJT, Watts DC, Wilson NHF, Wilson MA. Currentstatus and rationale for composite inlays and onlays. Br DentJ 199t;t70:269-273.

1. Clinical Research Associates" Use survey—1990. Clin Res As-soc Newsletter. 199Ü;14(12):3.

.1. Burke FJT, Watts DC, Wilson NHF. l-racturc resistance ofpremolar teeth restored with MOD composite inlays, the effectof cavity wall taper [abstract 14]. J Dent Res 1990;69:956.

4. WattsDC. Burke FJT, Wilson NHF. The effect of cavity designfactors on fracture resistance of teeth restored with compositeinlays (submitted).

3. Vale WA. Cavity preparation and further thonghts on highspeed. Br Dent J 1959:107:333-346.

6. Burke FJT. Investigation of aspects of optimum cavity designfor composite inlays [thesis]. University of Manchester, 1991.

7. Hood JAA. Biomechanics of the intact, prepared and restoredtooth: Some chnicaí implications. Int Dent J 1991;41:25-32.

8. Eakle WS. Increased fracture resistance of teeth; comparisonof five bonded composite resin systems. Quintessence Int198ö,t7:t7-2u.

9. Eakle WS. Fracture resistance of teeth restored with Class IIbonded composite resin. J Dent Res 1986;65:t49-153.

10. Morin D, DcLong R. Douglas WH. Cusp reinforcement by theacid-etch technique. J Dent Res 1984;63:IO75-t078.

11. Salis SG, Hood JAA, Kirk EEJ, Stokes ANS. Impact-fractureenergy of human premolar teeth. J Prosthet Denl 19S7;58:43-4S.

12. Malcolm PJ. Cast Restorations and Cusp Flexibility [Thesis].University of Otago. New Zealand, 1973.

13. Malcolm PJ, Hood JAA. The effect of cast restorations in re-ducing cusp flexibility in restored teeth [abstract 67]. J DentRes 1977;56:207.

14. Fisher DW, Caputo AA, Shilhngburg HT, Duncanson MF.Photoelastic stress analysis of inlay atid onlay preparations.J Prosthet Dent 1975;33:47-53.

15. Reagan SE, Schwandt NW, Duncanson MG. Fracture resist-ance of wide-isthmus mesio-occlusal-distal preparations withand without amalgam cuspal coverage. Quintessence Int1989:20:469^72.

16. Lopes LMP, Leitao JGM, Douglas WH. Effect of a new resininlay/onlay restorative material on cuspal reinforcement. Quin-tessence Int t99t;22:641-645.

17. Wendt SL. Microieakage and cusp fracture resistance of heat-treated composite resin inlays. Am i Dent 1991,4:10-14. D

Quintessence International Volume 24, Number 12/1993