The effect of dentin bonding agents on the microleakage of porcelain veneers

Dent Mater 10:278-281, July, 1994

The effect of dentin bonding.agents on the mzcroleakage of porcelazn veneers Christina Sire 1, Jennifer Neo ~, Ee Kiam Chua 1, Bee Yian Tan s Department of Restorative Dentistry, Government Dental Clinic, Singapore 2Department of Restorative Dentistry, National University of Singapore, Singapore

ABSTRACT Objectives. This in vitro study was conducted to investigate the effect of dentin bonding agents on the microleakage of porcelain veneers cemented on dentin margins. Methods. Forty human premolars were prepared on both the buccal and palatal surfaces. Each half was randomly assigned to eight groups. The veneer preparation of one group was cut entirely within enamel, to serve as controls. The remaining groups had cervical margins extending 1 mm beyond the cemento-enamel junction. Six dentin bonding agents were tested. The specimens were subjected to thermocycling and evaluated for marginal leakage using a silver nitrate stain technique. After sectioning, the extent of microleakage was measured at four interfaces: (1) incisal porcelain-composite; (2) incisal tooth-composite; (3) cervical porcelain-composite; and (4) cervical tooth-composite. Results. Statistical analysis showed that the cervical toothcomposite interface had a significantly higher leakage value (p<0.05) than the other three interfaces when the cervical margins were in dentin. Significance. The dentin bonding agents tested did not significantly reduce the marginal leakage of the porcelain veneers cemented on dentin margins. INTRODUCTION The development of porcelain veneers as a conservative treatment to mask discolored teeth and to restore fractured or malaligned teeth has proved to be a boon to the practice of dentistry. Various clinical and laboratory methods of fabricating veneers have been extensively described in the literature (Horn, 1983a; Calamia, 1983). Most studies advocated tooth preparation with a minimal reduction of 0.5 mm to ensure its placement within enamel (Horn, 1983b; Calamia, 1985). Studies evaluating the marginal integrity of porcelain veneers showed large marginal openings. Sire and Ibbetson (1993) reported veneer margin openings of 60-140 ~tm while Sorensen et al. (1992) measured marginal discrepancies in the range of 187-292 ~tm. Adequate bonding between tooth and composite and between porcelain and composite is necessary to eliminate microleakage. Excellent bond strengths between porcelain and composites have been well documented (Simonsen and Calamia, 1983; Calamia and Simonsen, 1984; Calamia et al., 1985; Stangel et al., 1987). Etching of enamel produces high bond strengths 278 Sim et aL/Effect of dentin bonding agents on porcelain veneers

with resin composites and significantly reduces marginal microleakage between the resin and enamel (Ortizet al., 1979). However, under certain conditions where the presence of root recession, caries, abrasion cavities or esthetic consideration necessitates the placement of cavosurface margins on dentin or cementum, an increased potential for microleakage, marginal staining, and recurrent caries can develop at these margins. This study was designed to determine the effectiveness of dentin bonding agents in reducing microleakage of porcelain veneers cemented on dentin margins.

MATERIALS AND METHODS Forty freshly extracted human premolars were collected, cleaned and stored in 1% thymol solution. The teeth were selected on the basis of sound tooth structure, absence of cracks and other defects. They were prepared on both the buccal and palatal surfaces; each half was randomly assigned to 8 experimental groups of 10 specimens each. The tooth preparation consisted of a labial reduction of 0.5 mm made freehand using depth orientation grooves formed by a diamond bur (HiDi 501, Claudius Ash, London, UK) and finished with white stones. A new bur was used after every fifth tooth preparation. One group acting as a positive control was prepared entirely within enamel with the cervical margin 1 mm coronal of the cementoenamel junction. The other seven groups had the cervical margins in root dentin 1 mm beyond the cemento-enamel junction. To act as a negative control, no dentin bonding agent was applied in one group. The six dentin bonding agents evaluated are listed in Table 1. Additional silicone impressions (Reprosil, Caulk/Dentsply, Milford, DE, USA) were taken of the tooth preparations and refractory dies cast (Refractory Material For Ceramics, Austenal Dental Inc., Chicago, I1, USA). A maximum of four firings was allowed for porcelain build-up (Microbond, Austenal Dental Inc., Chicago, I1, USA) and this included the glazing process. Porcelain veneers which showed cracks or any other defects were remade. The refractory die material was carefully removed from the porcelain veneers using a tungsten carbide bur until only a thin layer of die material was left. This layer was then removed by airblasting with 50 ~xn aluminium oxide at ~70 Pa. The porcelain veneers were etched for 1 rain with hydrofluoric acid (Porcelain Etch, Ultradent Products Inc, Salt Lake City, UT, USA) and then soaked in sodium bicarbon-

tions. The porcelain veneers were gently cemented onto the teeth using a light-cured Dentin Bonding Components Manufacturer resin composite (Porcelite, Kerr, Romulus, Agents MI, USA). Excess marginal resin was removed with a sable brush wetted with unBayer, Leverkusen, Gluma 'Cleanser'-EDTA Bis-GMA Germany 'Bond'-HEMA TEGDMA filled resin and the luting resin cured for gluteraldehyde 60 s with a visible light curing unit (MaxLite, Caulk/Dentsply, Milford, DE, USA). 3M, St Paul, MN, Scotchbond 2 Maleic acid, HEMA Bis-GMA To simulate clinical conditions, finishing of HEMA USA the veneers was done using diamond finishing burs (Micron Finishing System, PreTenure Aluminum oxalate Soln A: NTG-GMA Den Mat, Santa Maria, mier Dental Products, Norristown, PA, USA) in nitric acid Soln B: PMDM CA, USA followed by a diamond polishing paste (Brilliant, Renfert GmbH, Hilzingen, Germany) ICI PLC, Macclesfield, Tripton Polyhexanide MPDM, applied with a webbed rubber cup. TEGDMA Cheshire, UK urethane The bonded specimens were stored in demethacrlyate distilled water at room temperature for lwk. They were then subjected to 200 thermoCaulk/Dentsply, Milford, PENTA Universal Bond 2 HEMA cycles between 5°C and 55°C with a 1 min PENTA gluteraldehyde DE, USA dwell time in each water bath. Following the thermocyling period, the apices of the Kerr, Romulus, XR B o n d Phosphonated phosphonated teeth were sealed with acrylic resin and the MI, USA demethacrlyate ester demethacrlyate teeth coated with two layers of nail varnish ester, to within 1 mm of the margin of the restoraUDMA tions. The teeth were placed in 50% aquealipathic demethacrlyate ous silver nitrate solution for 2 h in the absence of light and then thoroughly rinsed with distilled water. Specimens were placed in a photo developing solution under a fluorescent light for 6 h, which caused areas of silver nitrate solution to turn black. The teeth were then sectioned into buccal and palatal halves. Each half was fixed onto a mounting plate with sticky wax and sliced longitudinally into three sections, each cut being made 0.5 mm on each side of the mid-plane using a 0.1 mm diamond C V wire saw (Wells Diamond Wire Saw, Model 3241-4, Well Walther Ebner, Switzerland) running at medium speed under a water coolant. Three specimens fractured during sectioning had to be discarded. The degree of microleakage of each 1 specimen was determined by the extent of penetration of silver nitrate stain from the cavosurface margin along the porcelaincomposite and tooth-composite interfaces (Fig. 1). Measurements of depth of silver nitrate penetration were made on both sides of the middle section under 40x magnification using a stereomicroscope equipped with a measuring device (MA 150, Fig. I. Schematic diagram of longitudinal section of specimen at cervical Meiji Labax Co Ltd., Tokyo, Japan). Readings were made at interface. Measurementof leakage along I- veneer.composite interface, 2- tooththe four interfaces: (1) cervical tooth-composite; (2) cervical composite interface, C- composite luting agent, T. tooth, V. veneer. porcelain-composite; (3) incisal tooth-composite; (4) incisal porcelain-composite, and recorded in millimeters. ate solution to neutralize the excess hydrofluoric acid. Comparison of the means for both the experimental groups The etched porcelain surfaces were then conditioned and the location sites were conducted by analysis of variance (Porcelain Liquid Acid Etch, Kerr, Romulus, MI, USA) and (ANOVA) using the SAS statistical software package (SAS Institute, Inc., Carey, NC, USA). The statistical significance of silane-treated (Porcelain Primer, Kerr, Romulus, MI, USA). The tooth specimens were cleaned with pumice and water differences between specific groups was determined using the and rinsed. To ensure that all preparations were in enamel Duncan's Multiple Range test. except for the cervical margins of the 70 specimens, a dentin RESULTS detector gel (Caulk/Dentsply, Milford, DE, USA) was used according to manufacturer's instructions on all specimens to The means and standard deviations of leakage values in each detect any dentin margins. Only the enamel surfaces were group were presented in Table 2. The positive control group etched with 37% phosphoric acid gel and then copiously rinsed showed excellent results with minimal leakage (Fig. 2) while with water spray. All dentin bonding agents were applied to cervical margins finished on dentin surfaces showed varied amounts of microleakage (Fig. 3). the tooth specimens according to the manufacturers' instrucTABLE 1: DENTIN BONDING AGENTS USED

Dental Materials~July 1994 279

TABLE 2: MEAN (+SD) LEAKAGE VALUES OF VENEERS AT FOUR INTERFACES (in mm) Group

Cervical Interfaces Tooth - Composite

Porcelain- Composite

Incisal Interfaces Tooth - Composite

Porcelain- Composite

Positive Control (n=10)

0.08+0.10 A'a

0.01_+0.02A,a

0.05+0.07 A,a

0.02+0.05 A,a

Negative Control (n=9)

0.72+0.62 B'a'b

0.01+0.02 A,a

0.02+0.04 A'a

0.00+0.00 A,a

Gluma (n=10)

0.37+0.23 B'a

0.00-+0.00 A'a

0.04+0.08 A'a

0.01 +0.02 A'a

Scotchbond 2 (n=10)

1.34+0.85 T M

0.00+0,00 A'a

0.14+0.23 A'a

0.07_+.0.16A'a

Tenure (n=9)

0.76+0.79 B'a'b

0.03+0.06 A'a

0.08+0.10 A'a

0.02+0.04 A'a

Tripton (n=9)

1.62+1.27 e,c

0.03+0.07 A'a

0.04+0.05 A'a

0.01+_0.02A'a

Universal Bond (n=10)

0.42+0.40 s'a

0.02+-0.05 A'a

0.03+0.04 A'a

0.01+0.02 A'a

XR Bond (n=10)

0.83+1.02 g ' a ' b

0.04+0.11A'a

0.04+0.08 A'a

0.01+0.03 A'a

Superscript letters indicate values that are not significantly different. (p < 0)

Fig. 3. Longitudinal section of specimen from Scotchbond 2 group at cervical tooth-composite interface. C- composite luting agent, D. dentin, V- veneer, 1veneer-composite interface, 2- dentin-composite interface.

(Table 3) showed that none of the dentin bonding agents significantly reduced microleakage as compared to the negative control group. It is interesting to note that Tripton showed a significant increase in leakage value (p<0.05). Fig. 2. Longitudinalsection of specimen from positive control group at cervical tooth-composite interface. C- composite luting agent, D- dentin, E. enamel, Vveneer, I- veneer-composite interface, 2- enamel-composite interface.

There was no significant difference among location sites for the positive control group (p>0.05) but significant differences were seen in the other seven groups (Table 2). The Duncan's Multiple Range test showed statistical differences between the cervical tooth-composite interface and the other three interfaces (p<0.05) for all groups except the positive control group. Comparison of group means for each site (Table 2) demonstrated a highly significant difference among groups only at the cervical tooth-composite resin interface. Analysis of the results at the cervical dentin-composite interface for the seven groups with cervical margins in dentin 280 Sim et aL/Effect of dentin bonding agents on porcelain veneers

DISCUSSION

The cervical enamel-composite and dentin-composite interfaces showed mean leakage values of 0.08 mm and 0.72 mm, respectively. Tjan et al. (1989) had mean leakage values of 0.12 mm and 1.8 mm for the corresponding interfaces. Lacyet al. (1992) measured mean gingival leakage values of 0.51 mm (Gluma), 0.39 mm (Tenure) and 1.92 mm (Scotchbond 2) compared to mean values of 0.37 mm (Gluma), 0.76 mm (Tenure) and 1.34 mm (Scotchbond 2) observed in this study. The data in this study concurred with the other studies on the inability of dentin bonding agents to effectively eliminate microleakage of porcelain veneers cemented on dentin margins. It has been widely reported that the bond between etched

TABLE 3: COMPARISON OF GROUP DIFFERENCES AT THE CERVICAL DENTIN - RESIN COMPOSITE INTERFACE USING DUNCAN'S MULTIPLE RANGE TEST Group

Leakage value + S.D. (mm)

Gluma

0.37 _+ 0.23a

Universal Bond 2

0,42 + 0.40 a

Negative Control

0.72 + 0.62b

Tenure

0.76 + 0.79 a'b

XR Bond

0.83 +_ 1.02 a'b

Scotchbond 2

1.34 + 0.85b,c

Tripton

1.62 + 1.27c

Superscript letters indicate no difference among groups at p < 0.05.

silanated porcelain and resin composite is predictable and strong (Simonsen and Calamia, 1983; Calamia and Simonsen, 1984; Calamia et al., 1985, Stangel et al., 1987). Also considering that the direction of polymerization contraction is towards the curing light source at the porcelain surface, disruption of the luting cement from the weaker dentin margin is likely. The bond between dentin and composite has been shown to be less predictable than that between enamel and composite resin. Many dentin bonding agents have been developed to improve the dentin-composite bond with a resultant decrease in microleakage (Kanca, 1989; Barkmeier and Cooley, 1989; Airoldi et al., 1992). However, despite using dentin bonding agents, no significant decrease in microleakage was seen this study. A possible reason could be due to the use of a dentin detector gel to stain the dentin. The gel consisted of an organic dye that reacted with the collagen protein in dentin. This reaction could possibly affect the subsequent bonding of the porcelain veneers to dentin surfaces. At present, there are no published studies on the effect of dentin detector gels on the dentin-composite bond. The purpose of using a dentin bonding agent was to pretreat the dentin surface to promote wetting of the resin composite. Since the dentin bonding agents did not show any decrease in microleakage, it would seem that the bonding agents tested did not markedly improve the wettability of the low viscosity composite luting agent. Ideally, a composite used to cement porcelain veneers should have a slightly lower viscosity than a composite used as a restorative material to facilitate easier handling of the veneers during cementation. As marginal infidelity and microleakage of porcelain veneers are unavoidable (Sim and Ibbetson, 1993; Sorensen et al., 1992), a probable solution to reduce microleakage is to seal all finished veneer margins with an unfilled resin as suggested by Zaimo~lu and Karaa ~aclio~glu (1991). Within the limits of this investigation, the study showed the cervical dentin-composite interface of resin-bonded porcelain veneers as the site most susceptible to marginal microleakage. None of the dentin bonding agents tested showed significant reductionin microleakage values. Instead, Tripton had a significant increase in microleakage when

compared to the negative control group. Much research is still needed to improve the reliability of dentin bonding agents and to achieve consistent results. Received May 4, 1992/Accepted May 15, 1994 Send reprint requests and correspondence to: Christina Sire Department of Restorative Dentistry Government Dental Clinic First Hospital Avenue Singapore 0316 Singapore

REFERENCES

Airoldi RL, Krejci I, Lutz F (1992). In vitro evaluation of dentinal bonding agents in mixed Class V cavity preparations. Quint Int 23:355-362. Barkmeier WW, Cooley RL (1989). Resin adhesive systems'In vitro evaluation of dentin bond strength and marginal microleakage. J Esthet Dent 1:67-72. Calamia JR (1983). Etched porcelain facial veneers: A new treatment modality based on scientific and clinical evidence. N Y J Dent 53:255-259. Calamia JR (1985). Etched porcelain veneers: The current state of the art. Quint Int 16:5-12. Calamia JR, Simonsen RJ (1984). Effect of coupling agents on bond strength of etched porcelain. JDent Res 62:297, Abstr. No. 1154. Calamia JR, Vaidyanathan J, Vaidyanathan TK, Hirsch SM (1985). Shear bond strength of etched porcelains. J Dent Res 64:296, Abstr. No. 1096. Horn HR (1983a). A new lamination: Porcelain bonded to enamel. N Y State Dent J 49:401-403. Horn HR (1983b). Porcelain laminate veneers bonded to etched enamel. Dent Clin North Am 27:671-684. Kanca J (1989). Microleakage of five dentin bonding agents. Dent Mater 5:415-416. Lacy AM, Wada C, Du W, Watanabe L (1992). In vitro microleakage at the gingival margin of porcelain and resin veneers. J Prosthet Dent 67: 7-10. Ortiz RF, Phillips RW, Swartz ML, Osborne JW (1979). Effect of composite resin bond agent on microleakage and bond strength. J Prosthet Dent 41:51-57. Sim C, Ibbetson RJ (1993). Comparison of fit of porcelain veneers fabricated using different techniques. Int J Prosthodont 6:36-42. Simonsen RJ, Calamia JR (1983). Tensile bond strength of etched porcelain. J Dent Res 62:297, Abstr. No. 1154. Sorensen JA, Strutz JM, Avera SP, Materdomini D (1992). Marginal fidelity and microleakage of porcelain veneers made by two techniques. J Prosthet Dent 67:16-22. Stangel I, Nathanson D, Hsu CS (1987). Shear bond strength of the composite bond to etched porcelain. J Dent Res 66:1460-1465. Tjan AHL, Dunn JR, Sanderson IR (1989). Microleakage patterns of porcelain and castable ceramic laminate veneers. J Prosthet Dent 61:276-282. Zaimo~lu A, Karaa~aclio~lu (1991). Microleakage in porcelain laminate veneers. J Dent 19:369-372. Dental Materials~July 1994 281