610 THE JOURNAL OF PROSTHETIC DENTISTRY VOLUME 81 NUMBER 5

Microleakage is the movement of bacteria, fluids,molecules or ions, and even air between the preparedcavity wall and the subsequently applied restorativematerials.1 Cervical lesions due to caries, erosion, orabrasion often have both enamel and dentin or cemen-tum margins. The longevity of a conventional Class Vrestoration can be affected by mechanical, thermal, and

chemical factors that result in stress in the cervicalarea.2,3

Bonded composites have been the common choicefor the esthetic restoration of Class V lesions. Howev-er, one disadvantage of composites is polymerizationshrinkage, which can result in marginal discrepanciesleading to microleakage, among other disadvantages.4This shrinkage has clinical repercussions such as sensi-tivity, marginal discoloration, and secondary caries.2,3

Many new bonding agents and glass ionomerrestorative materials have been introduced to bondrestorative materials to dentin and cementum marginsof cervical lesions,5,6 but microleakage at the dentin(cementum) aspects of restorations remains a problemof clinical significance.1,4,5 Glass ionomers are alterna-

Microleakage of Class V resin-modified glass ionomer and compomerrestorations

Manuel Toledano, MD, BDS, PhD,a Estrella Osorio, LDS, PhD,b Raquel Osorio, LDS, PhD,c andFranklin García-Godoy, DDS, MSd

University of Granada, Granada, Spain, and University of Texas Health Science Center at San Antonio,San Antonio, Texas

Statement of problem. Resin-modified glass ionomers and polyacid-modified resin composites (com-pomers) have been introduced to provide esthetic restorations. However, there is concern about the margin-al sealing ability of these materials, especially at the dentin (cementum) aspects of restorations.Purpose. This in vitro study evaluated the microleakage of Class V restorations made with resin-modifiedglass ionomers or a compomer.Material and methods. Thirty noncarious human molar teeth were used. Standardized kidney-shapedClass V cavity preparations were placed in the buccal and lingual surfaces at the cementoenamel junction.Teeth were randomly assigned to 3 experimental groups of 10 teeth each and restored as follows: group 1,Fuji II LC; group 2, Vitremer; and group 3, Dyract. In all cases, the manufacturers’ instructions were strict-ly followed. All materials were placed in a single increment. Unfinished restorations were immediately coat-ed with the respective manufacturers’ sealer or varnish and this was either light cured for 20 seconds orallowed to air-dry. After 24 hours, teeth were finished to contour and to the cavosurface margins, coatedwith nail varnish except for 1 mm around the restoration margin, thermocycled (1000×, 5-55°C) and placedin a solution of 2% basic fuchsin dye for 24 hours at room temperature. The staining along the toothrestoration interface was recorded.Results. Kruskal-Wallis 1-way analysis of variance revealed significant differences among all restorativematerials for the overall, occlusal, and gingival scores (P=.03, P=.01, P=.01, respectively). Occlusal and gin-gival scores for each matched pair of restorative materials using the Wilcoxon test showed statistically signifi-cant differences between Fuji II LC glass ionomer cement and Dyract composite, both for the occlusal(P=.005) and gingival (P=.005) margins and also as an overall evaluation (P=.01), with Fuji II LC showingthe least dye penetration. Vitremer revealed dye penetration scores not significantly different from Fuji IILC glass ionomer cement or Dyract composite.Conclusion. Resin-modified glass ionomers showed less or similar microleakage than the polyacid-modi-fied composite resin tested. (J Prosthet Dent 1999;81:610-5.)

aProfessor, Department of Dental Materials, University of Granada,Spain.

bAssistant Professor, Department of Dental Materials, University ofGranada.

cProfessor, Department of Dental Materials, University of Granada.dProfessor and Director of Clinical Materials Research, Department

of Restorative Dentistry, Dental School, University of TexasHealth Science Center at San Antonio.

CLINICAL IMPLICATIONS

The findings of this in vitro study suggest that Fuji II LC glass ionomer cement pro-vided a better marginal adaptation than Dyract composite in Class V restorations,and the amount of resin content and filler particles of the materials may influence thedegree of microleakage.

tive materials to composites for the conservativerestoration of these lesions because of their adhesion totooth structure, fluoride release, biocompatibility,lower shrinkage values, reduced microleakage, andacceptable esthetics.7-11 Light-cured resin-modifiedglass ionomer cements were developed to improve thehandling and working characteristics of the originalglass ionomer formulation.12,13 Improved adhesion todentin is probably caused by both a chemical bondfrom the polyacrylic acid component and formation ofa hybrid layer from the hydrophilic HEMA.14-21

Favorable adhesive and fluoride-releasing propertiesof glass ionomer cements have lead to their widespreaduse as restorative, lining, and luting materials. To over-come the problems of moisture sensitivity and low earlymechanical strengths associated with the conventionalglass ionomer cements (GICs) and at the same timemaintain their clinical advantages, some hybrid versionsof GIC were introduced that are light-cured, becauseof their small quantity of resin components such asHEMA or BIS-GMA. In some situations, the polyacidalso has been modified with side chains that can bepolymerized by light-curing mechanisms. The actualformulations vary between manufacturers, but theamount of resin in the final set restoration is between4.5% to 6%, such as for Fuji II LC and Vitremer glassionomer cements. The addition of a resin componentto GIC and its effects on the development of the ioniccrosslink and the subsequent marginal seal against thetooth structure needs further evaluation.

To overcome technique-sensitive mixing and han-dling properties of the resin-modified glass ionomercements, new materials containing acid-decomposableglass and acidic polymerizable monomers substitutingthe polyalkenoic acid polymer were developed. Thesematerials were termed polyacid-modified resin compos-ites,22-26 commonly called compomers. Dyract poly-acid-modified resin composite belongs to the newmaterials that have either been marketed as multipur-pose materials, or contain both of the essential compo-nents of a glass-ionomer cement but at levels that areinsufficient to produce an acid-base reaction.13 Withthis material, the resin content is approximately 28%.

The purpose of this study was to compare themicroleakage of Class V restorations produced with the3 materials, which differ in their resin content, to testthe hypothesis that resin content affects microleakage.

MATERIAL AND METHODS

Thirty noncarious human molars, which were storedin a solution of 1% sodium hypochlorite for up to4 months at room temperature, were test specimens.After surface debridement with a hand-scaling instru-ment and cleaning with a rubber cup and slurry ofpumice, a standardized Class V cavity preparation wasplaced in the buccal and lingual surface at the cemen-

toenamel junction. Preparations were made with ano. 329 carbide bur in a high-speed handpiece and atemplate to a uniform kidney-shaped outline. Prepara-tions measured 5 mm long, 3 mm wide, and 2 mmdeep with the occlusal margin in enamel and the gingi-val margin in dentin or cementum.

Subsequently, teeth were randomly assigned to3 experimental groups of 10 teeth each. Buccal and lin-gual preparations of group 1 were restored with Fuji IILC (GC Corp, Tokyo, Japan) resin-modified glassionomer cement; group 2 with Vitremer (3M, St Paul,Minn.) resin-modified glass ionomer cement; andgroup 3 with Dyract (De Trey Dentsply, Konstanz,Germany) polyacid-modified resin composite.

In all cases, the manufacturers’ instructions fordentin conditioning, powder/liquid proportioning andmixing were strictly followed. For Fuji II LC glassionomer cement, the cavity wall was conditioned for20 seconds with dentin conditioner (GC Dental Corp).For Vitremer glass ionomer cement, Vitremer primer(3M Dental Products) was applied on the cavity wallfor 20 seconds, gently air dried and light cured for30 seconds. For Dyract composite restorations, the cav-ity wall was treated with PAS primer/adhesive (DeTreyDentsply) for 30 seconds, excess was removed with ablast from an air syringe and the adhesive was cured for20 seconds. A second coat of the primer/adhesive wasapplied and immediately light cured for 20 seconds.Dyract composite was placed in 1 increment. The teethwere prevented from dehydration by remaining indeionized water storage at room temperature when notbeing prepared for restoration.

Immediately after the restorative material was placed,a clear cervical matrix (Clear Thru, Premier DentalProducts, Norristown, Pa.) was adapted over the resin-modified GIC restorations and the materials cured witha visible light source (Optilux 400, Demetron ResearchCorp, Danbury, Conn.) in accordance with the manu-facturers’ recommended time. The light was tested forlight output (>600 mW/cm2) before each use with aDemetron radiometer (model 100, Demetron ResearchCorp). When the matrix was removed, the unfinishedrestorations were immediately coated with the respectivemanufacturer’s sealer or varnish, and this was eithercured with a visible light source or allowed to dry beforereturning the tooth to deionized water storage at roomtemperature. After 24 hours, the teeth were finished tocontour and to the cavosurface margins with a no. 7901carbide finishing bur (SS White, Lakewood, N.J.) withair and water spray in a high-speed handpiece (Star Den-tal, Lancaster, Pa.) and medium, fine, and super fine Sof-Lex disks (3M Dental Products), which were first lubri-cated with water and used in sequence with air-waterspray in a slow-speed handpiece (Star Dental). Theapical portions of all specimens were filled ad retrumwith a zinc oxide-eugenol paste.

TOLEDANO ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

MAY 1999 611

Teeth were prepared for microleakage evaluation bycoating the entire tooth with 1 application of nail var-nish, except for 1 mm around the restoration margins.These specimens were then subjected to 1000 temper-ature cycles as suggested in a previous study.27 Eachcycle consisted of 30 seconds at 6°C and 30 seconds at60°C. After thermocycling, teeth were placed in a solu-tion of 2% basic fuchsin dye (Fisher Scientific, FairLawn, N.J.) for 24 hours at room temperature.

After removal of the specimens from the dye solu-tion, the superficial dye was removed with a pumiceslurry and rubber cup. Teeth were then mounted in alight-curing 1-component methacrylate-based resin(Technovit 7200 VLC, Kulzer, Norderstedt, Germany)to facilitate handling during sectioning. The resin wascured for 24 hours (Histolux, EXAKT, Norderstedt,Germany), then teeth were sectioned longitudinallywith a hard tissue microtome (Exakt-apparerteban,Otto Herrman, Norderstedt, Germany) in 0.6-mmthick sections to evaluate the dye penetration.28 Thesections were then separated, and the cut surfaces cor-responding to the most mesial, central (mesial and dis-tal), and most distal portion of the tooth restorationinterface were examined at the occlusal and gingivalmargins with a stereomicroscope (Olympus Co, Tokyo,Japan) at ×16 magnification. Examination of the speci-mens was undertaken at random, and the investigatorswere unaware of the exact nature of the restorativematerial.

Staining along the tooth restoration interface wasrecorded by 2 evaluators, according to the followingcriteria: 0 = no dye penetration; 1 = partial dye pene-tration; 2 = dye penetration along the occlusal orgingival wall, but not including the axial wall; and 3 =dye penetration to and along the axial wall. If disagree-ment occurred between the evaluators, a consensus wasobtained after reexamination of the specimen by bothinvestigators. Occlusal, gingival, and overall scores foreach group of restoration were compared with theKruskal-Wallis 1-way analysis of variance (ANOVA)nonparametric statistical test to identify any statistical

significant differences between the materials, and theWilcoxon test was performed to compare each matchedpair of restorative materials. Significance was consid-ered at the .05 level.

RESULTS

Microleakage scores for the occlusal, gingival, andoverall walls are presented in Table I. Kruskal-Wallis1-way ANOVA indicated significant differencesbetween the restorative materials for overall, occlusal,and gingival scores (P=.03; P=.01; P=.01, respectively).Further matched analysis by Wilcoxon test was under-taken to compare occlusal, gingival, and overall scoresof each material, which revealed statistically significantdifferences between Fuji II LC glass ionomer cementand Dyract resin composite, both for the occlusal(P=.005) and gingival (P=.005) margins and also as anoverall evaluation (P=.01) (combining the occlusal andgingival margins scores) with Fuji II LC demonstratingthe least dye penetration between these 2 products.Vitremer glass ionomer cement revealed dye penetra-tion scores between Fuji II LC glass ionomer cementand Dyract resin composite, with no statistically signif-icant differences between Vitremer glass ionomercement and the other 2 products.

DISCUSSION

Polymerization shrinkage of resin-containing restora-tive materials may result in marginal discrepancies thatlead to microleakage, marginal discoloration, and sensi-tivity.2-4 Hygroscopic expansion can compensate, tosome degree, for polymerization shrinkage. Water sorp-tion can help to reduce marginal gaps3; for this reason,glass ionomer cements, which absorb the most waterduring the first 24 hours after placement,6 can displayless microleakage than resins. Attin et al7 reported thatFuji II LC glass ionomer cement expanded after curingand immersion in water, whereas Dyract resin compositeand Vitremer glass ionomer cement revealed a total vol-umetric loss. Thus, they concluded that water expansionis 1 factor that reduces the leakage.

THE JOURNAL OF PROSTHETIC DENTISTRY TOLEDANO ET AL

612 VOLUME 81 NUMBER 5

Table I. Microleakage of the different groups

Occlusal Gingival Overall

Group 0 1 2 3 0 1 2 3 0 1 2 3

Fuji II LC (n = 17) 14 1 0 2 12 2 0 3 10 3 0 4Vitremer (n = 18) 9 3 2 4 8 1 4 5 4 4 4 6Dyract (n = 22) 7 4 6 5 6 3 0 13 5 3 1 13

Kruskal-Wallis 1-way analysis of variance indicated significant differences between all the restorative materials for both overall, occlusal, and gingival scores(P=.03; P=.01; P=.01, respectively).Wilcoxon test (to compare occlusal, gingival, and overall scores of each material) revealed that the occlusal and gingival scores for each matched pair of restora-tive materials showed statistically significant differences between Fuji II LC and Dyract, both for the occlusal (P=.005) and gingival (P=.005) margins and also asan overall evaluation (P=.01) with Fuji II LC showing the least dye penetration. Although Vitremer revealed dye penetration scores between Fuji II LC and Dyract,there were no statistically significant differences between them. Also, there were no statistically significant differences between Dyract and Vitremer.

Our results disagree with those of Yap et al8 whocompared the microleakage of Dyract resin compositeand Fuji II LC glass ionomer cement and reported nostatistically significant differences in microleakagescores. In their study, they reported a significant differ-ence between enamel and dentin; in our study, even ifmicroleakage was less common in enamel, the differ-ence was not significantly different. These differencesbetween the studies could be because Yap et al8 storedtheir specimens in a saline solution for 1 week beforetesting. This storage time allows hygroscopic expansionof the material,7 which may compensate the originalpolymerization shrinkage of the material, which allowsless microleakage. In our study, specimens were ther-mally cycled for approximately 2 days, and the materialmay not have expanded completely. Yap et al8 also sug-gested that 1 of the unique features of the resin thatreleases fluoride to enamel is the omission of acid etch-ing, which is a critical step in most resin composite andadhesive systems. The manufacturers have claimed thatthis is achieved through the use of a specially formulat-ed coupling agent with hydrophilic phosphate groupsthat is thought to form ionic bonds with the calcium ofhydroxyapatite. Dyract resin composite also aims to beself-adhesive because of hydrophilic carboxylic groupspresent in its patent tetrachlorobiohenyl (TCB) resin.These questions need further investigation.

No restorative material evaluated in our study com-pletely resisted microleakage at the occlusal or gingi-val walls of the tooth. Of the 3 products evaluated,Fuji II LC glass ionomer cement exhibited the leastdye penetration, at both the occlusal and gingivalmargins, and when evaluated as overall values (enam-el and gingival scores pooled together). However,only with the overall evaluation did Fuji II LC glassionomer cement reveal a statistically significant differ-ence with Dyract resin composite. The lack of statisti-cally significant difference in microleakage betweenresin-modified glass ionomers has also been previous-ly reported.15 Uno et al16 concluded that the superioradaptation of Fuji II LC glass ionomer cement to thecavity walls was responsible for the lower dye penetra-tion, which may be a result of the glass ionomercement undergoing minimal setting shrinkage over alonger period and approximately one half that ofresins.17 Because the resin component is responsiblefor the polymerization shrinkage, and Dyract resincomposite has more resin than Fuji II LC glassionomer cement in its composition, it is possible thatthis is the reason for the greater microleakage scoresobserved with Dyract resin composite. Another reasonthat could explain the results is the resin componentof Fuji II LC glass ionomer cement undergoing dif-ferent rates of polymerization shrinkage during lightcuring (as it is a dual-cure material) compared withDyract resin composite.

The microleakage scores for Vitremer glass ionomercement fell between those recorded for Dyract resincomposite and Fuji II LC glass ionomer cement, whichcould be due to 2 reasons. Fuji II LC is a resin-modifiedglass ionomer in which the HEMA content is merelyblended with a polyalkenoic acid liquid, whereas Vit-remer, in addition to being a simple mixture of HEMAwith polyalkenoic acid, is also modified by the attach-ment of polymerizable methacrylate side groups.13 It ispossible that Vitremer has more polymerizable resinthan Fuji II LC, but less than Dyract; its microleakagevalues fell in between these 2 materials. The better adap-tation of Fuji II LC glass ionomer cement comparedwith Dyract resin composite could be also due to the15-second dentin conditioning performed with the10% polyacrylic acid. This dentin treatment produces aclose relation between the ionomer and dentin struc-tures as it removes the smear layer, leaving the surfaceclean and theoretically better able to accept a glassionomer.18 Moreover, the Fuji II LC liquid containsapproximately 40% HEMA (manufacturer’s data) andprimers that contain similar hydrophilic monomers thanresin-containing materials, facilitating the bondingbetween dentin and these type of materials.19

Although the PAS adhesive of the Dyract resin com-posite tested had orthophosphoric acid to conditionthe dentin, it also contained TGDMA and elastomericresins, which have chemical affinity with the resin con-tained in the material. When these resins shrink duringpolymerization, they could generate a gap wheremicroleakage could be detected. The extent of the cur-ing shrinkage determines the formation of marginalgaps if the restorative material does not adhere enoughto tooth structure or it can cause cohesive failures inthe material.23

The application of Vitremer glass ionomer cementonly requires the primer application and light curing for20 seconds. It is possible that the pH of the dentinprimer could modify the smear layer sufficiently to per-mit the tooth and restorative material to come into inti-mate interfacial contact.18 Charlton and Haveman20

obtained higher bond strength values to dentin with FujiII LC glass ionomer cement than with VariGlass VLCresin. Some consider this latter material a light-curedglass ionomer and not a true polyacid-modified com-posite because it does not have an acid-base cure reac-tion.13 These differences in bond strength values couldcontribute, among other factors, to explain the differ-ences in the microleakage patterns recorded in ourstudy. Polymerization shrinkage also produces materialshrinkage in all directions and most often the dentinmargins are unprotected to resist microleakage.2 Withrestorations made with resin-reinforced glass ionomers,the adhesion is mainly due to a physicochemical reactionwith dentin and enamel due to the polar nature of thepolyacrylates and minerals for the dental hard tissues.

TOLEDANO ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

MAY 1999 613

However, this bonding is not so strong and does notproduce an adequate marginal sealing.20,21

The increase in leakage of Dyract resin compositealso could be attributed to thermal expansion mismatchwith tooth substance, which is reported to be signifi-cantly higher than that of conventional cements andalso less than that of composites,9,10 perhaps due totheir different chemical composition. Leakage of com-posite resin restorations may be attributed to a con-traction gap produced by polymerization shrinkage andexpansion and contraction with temperature changes,because the coefficient of thermal expansion of com-posites is different from that of the dental hard tissues.Glass ionomer cements exhibit limited shrinkage dur-ing setting and their coefficient of thermal expansion issimilar to that of dentin.4 Mitra and Conway9 reportedthat Fuji II LC and Vitremer materials had coefficientsof thermal expansion of 31.5 and 11.5 ppm/°C,respectively, and Silux Plus microfilled composite 56.6ppm/°C 7 days after curing. Dyract has a compositionclosely related to the microfilled composites and has acoefficient of thermal expansion of 40.52 ppm/°C(P Hammesfahr, verbal communication, 1998). Thismay explain why Dyract resin composite is more sus-ceptible to thermal stresses than the other materials.Also, because the resin component of the materialadheres poorly to the cervical dentin than to enamel,this justifies, in part, that the Dyract resin compositerevealed more leakage at the gingival margin than atthe enamel margin.

Although Vitremer glass ionomer cement displayedmicroleakage values between those of Fuji II LC andDyract materials, there was no statistically significantdifference among the 3 materials. Some authors havepointed out that significant dimensional changes andsurface hardening can occur after initial light curing ofthe resin component of resin-modified glass ionomers,and further contraction continues for the first 24 hoursas the material matures.10,11 Because both Vitremerand Fuji II LC glass ionomer cements contain approx-imately the same percentage of resin, which is less thanthat for Dyract composite, it could be thought that thisis another reason to explain the different microleakagepatterns.10,11 Uno et al16 considered that the differ-ences observed between Vitremer and Fuji II LC glassionomer cements might be due to differences in matu-ration of setting reactions.

Although the results obtained from this study maynot be directly extrapolated to the clinical situation,they provide some information regarding the perfor-mance of the restorative system evaluated. Independentlong-term clinical data are still required.

CONCLUSIONS

Within the limits of this study, the following conclu-sions were drawn:

THE JOURNAL OF PROSTHETIC DENTISTRY TOLEDANO ET AL

614 VOLUME 81 NUMBER 5

1. The resin-modified glass ionomers showed less orsimilar microleakage than the polyacid-modified com-posite resin tested.

2. The amount of resin content and filler particles ofthe materials may influence the degree of microleakage.

REFERENCES

1. Kidd EA. Microleakage: a review. J Dent 1976;4:199-206.2. Davidson CL. Resisting the curing contraction with adhesive composites.

J Prosthet Dent 1986;55:446-7.3. Feilzer AJ, de Gee AJ, Davidson CL. Relaxation of polymerization con-

traction shear stress by hygroscopic expansion. J Dent Res 1990;69:36-9.4. Kaplan I, Mincer HH, Harris EF, Cloyd JS. Microleakage of composite

resin and glass ionomer cement restorations in retentive and nonretentivecervical cavity preparations. J Prosthet Dent 1992;68:616-23.

5. Gordon M, Plasschaert AJM, Stark MM. Microleakage of several tooth-colored restorative materials in cervical cavities. A comparative study invitro. Dent Mater 1986;2:228-31.

6. Mount GJ. Atlas práctico de cementos de ionómero de vidrio. Barcelona:Salvat; 1990. p. 128.

7. Attin T, Buchalla W, Kielbassa AM, Helwig E. Curing shrinkage and volu-metric changes of resin-modified glass ionomer restorative materials.Dent Mater 1995;11:359-62.

8. Yap AU, Lim CC, Neo JC. Marginal sealing ability of three cervical restora-tive systems. Quintessence Int 1995;26:817-20.

9. Mitra SB, Conway WT. Coefficient of thermal expansion of somemethacrylate-modified glass ionomers. J Dent Res 1994;73:219 (abstr944).

10. Cárdenas HL, Burgess JO. Thermal expansion of glass ionomers. J DentRes 1994;73:220 (abstr 946).

11. Hallett KB, García-Godoy F. Microleakage of resin-modified glassionomer cement restorations: an in vitro study. Dent Mater 1993;9:306-11.

12. Antonucci JM, Mckinney JE, Stansbury JW. Resin-modified glass-ionomercement restoration. US Patent 160856, 1988.

13. Gladys S, Van Meerbeek B, Braem M, Lambrechts P, Vanherle G. Com-parative physico-mechanical characterization of new hybrid restorativematerials with conventional glass-ionomer and resin composite restora-tive materials. J Dent Res 1997;76:883-94.

14. Ferrari M, Davidson CL. Interdiffusion of a traditional glass ionomercement into conditioned dentin. Am J Dent 1997;10:295-7.

15. Sidhu SK. Sealing effectiveness of light-cured glass ionomer cement lin-ers. J Prosthet Dent 1992;68:891-4.

16. Uno S, Finger WJ, Fritz UB. Effect of cavity design on microleakage ofresin-modified glass ionomer restorations. Am J Dent 1997;10:32-5.

17. Trushkowsky RD, Gwinnett AJ. Microleakage of Class V composite, resinsandwich, and resin-modified glass ionomers. Am J Dent 1996;9:96-9.

18. Pachuta SM, Meiers JC. Dentin surface treatments and glass ionomermicroleakage. Am J Dent 1995;8:187-90.

19. Van Meerbeek B, Inokoshi S, Braem M, Lambrechts P, Vanherle G. Mor-phological aspects of the resin-dentin interdiffusion zone with differentdentin adhesive systems. J Dent Res 1992;71:1530-40.

20. Charlton DG, Haveman CW. Dentin surface treatment and bond strengthof glass ionomers. Am J Dent 1994;7:47-9.

21. Zyskind D, Frenkel A, Fuks A, Hirschfeld Z. Marginal leakage around V-shaped cavities restored with glass-ionomer cements: an in vitro study.Quintessence Int 1991;22:41-5.

22. McLean JW, Nicholson JW, Wilson AD. Proposed nomenclature for glass-ionomer dental cements and related materials. Quintessence Int 1994;25:587-9.

23. Van Dijken JW. 3-year clinical evaluation of a compomer, a resin-modi-fied glass ionomer and a resin composite in Class III restorations. Am JDent 1996;9:195-8.

24. Abdalla AI, Alhadainy HA, García-Godoy F. Clinical evaluation of glassionomers and compomers in Class V carious lesions. Am J Dent 1997;10:18-20.

25. García-Godoy F, Hosoya Y. Bonding mechanism of Compoglass to dentinin primary teeth. J Clin Pediatr Dent 1998;22:217-20.

26. El-Kalla IH, García-Godoy F. Bond strength and interfacial micromor-phology of compomers in primary and permanent teeth. Int J PaediatrDent 1998;8:103-14.

TOLEDANO ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

MAY 1999 615

27. Crim GA, García-Godoy F. Microleakage: the effect of storage and cyclingduration. J Prosthet Dent 1987;57:574-6.

28. Mixson JM, Eick JD, Chappell RP, Tira DE, Moore DL. Comparison of two-surface and multiple-surface scoring methodologies for in vitromicroleakage studies. Dent Mater 1991;7:191-6.

Reprint requests to:DR. FRANKLIN GARCIA-GODOY

DEPARTMENT OF RESTORATIVE DENTISTRY

UNIVERSITY OF TEXAS HEALTH SCIENCE CENTER

7703 FLOYD CURL DR

SAN ANTONIO, TX 78284-7850FAX: (210) 567-3522E-MAIL: godoy@uthscsa.edu

Copyright © 1999 by The Editorial Council of The Journal of ProstheticDentistry.

0022-3913/99/$8.00 + 0. 10/1/96694

Shear stresses in the adhesive layer under porcelain veneersTroedson M, Derand T. Acta Odontol Scand 1998;56:257-62.

Purpose. In vitro studies into which part of the enamel-resin–composite-porcelain laminate sys-tem breaks have shown that the luting interface is the weakest part of the lamination and that itwill fail due to sheer stresses. This study calculated sheer stress in the composite cement andenamel bond with the facing loaded in the incisal area under different angles and adhesive con-ditions.Material and methods. Two-dimensional finite element models of veneers on teeth with anintermediate layer of resin were designed according to the size of an average maxillary centralincisor. The abutment was considered to be homogenous, and the remaining enamel layer underthe buccal surface of the veneer and the pulp were treated as dentin with regard to material prop-erties. Three models of the tooth were created with different margin designs, while all designshad preparations that covered the incisal edge. Porcelain facings were made to be 0.5 mm thick:composite cement layer, 25 µm; enamel bond layer, 1 µm. Three adhesive conditions were test-ed: (1) lack of polymerization in the facing’s periphery, (2) lack of polymerization in the middle,and (3) total bonding of the facing. All models were loaded at 0, 30, and 60 degrees to the longaxis of the tooth.Results. Rather extensive tables presented the numeric results of the study. Maximum sheerstresses did not exceed the stress level for debonding, but great differences in maximum shearstress appeared with varying loss of bond and different loading angles. Fully laminated facingshowed stress levels in the composite cement to be only 1⁄5 of those in the facing with a lack ofadhesion in the periphery and 1⁄15 of those in the enamel bond. Maximum stresses increased about4 times when the load angle was 30 degrees compared with 0 degrees, and increased 1.5 timesfrom 30 to 60 degrees.Conclusions. A porcelain veneer that is kept inside the enamel, with full lamination, demon-strated fairly low shear stresses in the enamel bond and composite cement and thus should indi-cate good long-term prognosis. 13 References.—ME Razzoog

Noteworthy Abstractsof theCurrent Literature