- Email: [email protected]
The effect of various surface treatments and bonding agents on the repaired strength of heat-treated composites
The effect of various surface treatments and bonding agents on the repaired strength of heat-treated composites Cristina Lucena-Martín, PhD,a Santiago González-López, PhD,b and José Manuel Navajas-Rodríguez de Mondelo, PhDc Faculty of Dentistry, University of Granada, Granada, Spain Statement of problem. Some clinical situations may require the repair of a secondary polymerized or aged composite. The higher indirect resin conversion rate may prove to be a disadvantage if a repair procedure based on covalent bonding from unreacted methacrylate groups is attempted. Purpose. This study evaluated the effectiveness of different combinations of surface treatments and 2 bonding agents used to enhance heat-polymerized and aged composite repairs. Material and methods. Ninety Herculite XRV and 90 Heliomolar Radiopaque specimens were prepared and then postpolymerized and stored for 4 weeks. All composites were subjected to 1 of 9 treatment regimens that involved adding fresh composite onto a corresponding postpolymerized composite (Herculite/Herculite or Heliomolar/Heliomolar). The surfaces were treated with different combinations of air abrasion, phosphoric acid, hydrofluoric acid, acetone, Special Bond II, Heliobond, and Prime & Bond 2.0. Results. Surface treatment with air abrasion resulted in the strongest repairs; surface treatment with phosphoric acid resulted in the weakest repairs. Conclusion. The use of air abrasion and Prime & Bond 2.0 adhesive consistently improved the shear bond strength for both composites tested. (J Prosthet Dent 2001;86:481-8.)
CLINICAL IMPLICATIONS Among the protocols tested, surface treatment with air abrasion and a 1-component adhesive most effectively enhanced the repair of aged and heat-polymerized microhybrid and microfilled composites.
T
he rising patient demand for esthetic dental restorations coupled with concerns about the toxicity of amalgam has markedly increased the use of composites. Despite some recent improvements, their technique sensitivity leads to numerous failures in the clinical setting, especially when they are used in posterior teeth. The repair of composites that exhibit small fractures, staining, or wear may be a viable and less costly alternative to their complete replacement and would cause less pulpal trauma.1 The bond strength between increments of composite is equal to the cohesive strength of the material.2,3 However, if the composite has been contaminated, polished, processed in a laboratory (indirect composite restorations), or aged, the adhesion to a new composite is reduced to 25% to 80% of the original cohesive strength.4-7 Various methods have been reported to improve the reactivity of highly converted composites; these methaAssistant
Professor, Department of Pathology and Dental Therapy. Professor, Department of Pathology and Dental Therapy. cProfessor, Department of Pathology and Dental Therapy. bAssociate
NOVEMBER 2001
ods include acid etching,8 air abrasion,5,9 and the use of solvents and silanes.10 There is no consensus on the results obtained with the different procedures. Several studies have supported the efficacy of multifunctional 1-component adhesives and their bond to enamel and dentin.11 However, their efficacy for composite repairs has not been adequately studied. This study was designed to evaluate the efficacy of different mechanical and chemical procedures used to improve the bond strength of highly polymerized microhybrid and microfilled composites.
MATERIAL AND METHODS The composites chosen for this study were a hybrid composite containing a BIS-GMA matrix resin (Herculite XRV; Kerr Manufacturing, Romulus, Mich.) and a microfilled composite based primarily on urethane dimethacrylate (Heliomolar Radiopaque; Vivadent, Schaan, Liechtenstein). Ten specimens of Herculite XRV and Heliomolar Radiopaque composite were prepared as controls. Twenty aluminum molds consisting of a disk 20 mm in external diameter and 3 mm in depth with a central THE JOURNAL OF PROSTHETIC DENTISTRY 481
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 1. Aluminum mold.
LUCENA-MARTÍN ET AL
cavity 5 and 8 mm in diameter (top and bottom base, respectively) were used (Fig. 1). The mold was placed on a glass slide, and a single increment of composite was condensed in each by means of a plastic instrument. The resin was immediately light-polymerized for 40 seconds, with the tip of the light-polymerizing unit (Optilux 401; Demetron Research Corp, Danbury, Conn.) placed in contact with the surface of the mold (Fig. 2). Light-polymerization was repeated for 40 seconds from underneath the glass slide. A similar polymerization area was achieved for all specimens, as each was maintained the same distance from the light source. A silicon disk (20 mm in external diameter and 4 mm in depth, with a central cavity 5 mm in diameter) was placed on the aluminum mold, and composite was condensed into it in 2 increments: The first increment filled two thirds of the cylinder height, and the second increment filled it to the top. Each increment was light-polymerized for 40 seconds. The resulting specimens were submitted to a secondary postpolymerization cycle (Ivomat 500; Ivoclar AG FL-9494, Schaan, Liechtenstein) with a combined application of heat (120°C) and pressure (6 atmospheres) for 10 minutes, within a maximum of 6 hours.
Study groups
Fig. 2. Light-polymerization of composite.
Fig. 3. Slide placed over fresh composite.
482
One hundred eighty aluminum molds were divided into 2 groups (n = 90 each), 1 group for Herculite XRV and 1 group for Heliomolar Radiopaque composite. Each mold was placed on a glass slide. Resin was placed within the mold in a single increment, and the first slide then was covered with another glass slide (Fig. 3). The composite was polymerized through the top slide to achieve a glossy, smooth surface. Because the total height of the mold was 3 mm, it was lightpolymerized for an additional 40 seconds through the bottom slide. Before the study and before each specimen group was polymerized, the polymerizing light output was tested with a Cure-Rite light intensity meter (Efos Inc, Mississagwa, Ontario, Canada). This ensured that the light intensity was not less than 300 W/m2. After secondary polymerization (120°C per 6 atmospheres per 10 minutes), the specimens were cooled for 10 minutes and then placed in an ultrasound bath for 2 minutes to remove any surface debris left behind by the postpolymerization treatment. The specimens were stored in distilled water for 4 weeks at room temperature. The specimens of each group were randomly assigned to 1 of 9 study subgroups for different combinations of surface treatment and bonding agents (Table I). The treatment protocols involved the use of air abrasion, 37% orthophosphoric acid, 9.6% hydrofluoric acid, Special Bond II, 99% acetone, Heliobond, and/or Prime & Bond 2.0. VOLUME 86 NUMBER 5
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
Table I. Treatment applied to composite Composite
Herculite
Heliomolar
Group
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Air abrasion
H3PO4
+
+ +
HF
Acetone
SB II
+ +
+
+ + + + + +
+ + + + +
+ + +
+ + + + + +
P&B 2.0
+ + +
+ + + + + + + + +
Heliobond
+ +
+ +
+ + + +
H3PO4 = 37% phosphoric acid; HF = 9.6% hydrofluoric acid; SB II = Special Bond II; P&B 2.0 = Prime & Bond 2.0.
Air abrasion: A flow of 50 µm aluminum oxide particles was used at a pressure of 60 psi (Microetcher; Danville Engineering, Danville, Calif.). The orifice of the instrument was placed 1 cm from the composite surface until the glossy appearance of the composite turned to a matte finish. 37% orthophosphoric acid: The acid (Email Preparator GS; Ivoclar) was applied for 1 minute. The surface of the resin then was washed for 45 seconds and dried with compressed air. 9.6% hydrofluoric acid: The acid (Bite Perf, Malaga, Spain) was applied for 2 minutes. The specimen then was carefully washed and dried with compressed air. Special Bond II: The surface of the resin first was treated with orthophosphoric acid. After a thin layer of Special Bond II (Vivadent) was applied according to the manufacturer’s instructions, the surface was dried for 20 seconds and light-polymerized for 40 seconds. An adhesive resin was not used because Special Bond II already contains an adhesive. 99% acetone: A brush was dipped several times into the acetone solution (Panreac; Monfplet & Esteban SA, Barcelona, Spain), which was used to coat the surface of the composite for 1 minute. The surface then was lightly dried to evaporate the remaining solvent. Heliobond: One increment of Heliobond (Vivadent) was applied, thinned with an air syringe, and then light-polymerized for 10 seconds. Prime & Bond 2.0: The adhesive (Dentsply/DeTrey, Konstanz, Germany) was applied according to the manufacturer’s instructions. Regardless of the type of surface treatment, the postpolymerized composite was treated with 37% phosphoric acid for 15 seconds. After NOVEMBER 2001
the washing and drying steps, a generous layer of Prime & Bond 2.0 was applied, ensuring that the surface of the resin remained wet for 20 seconds. The remaining solvent was removed with compressed air, and the resin was light-polymerized for 10 seconds. A second increment of adhesive was applied, the excess solvent was removed, and the resin was again lightpolymerized for 10 seconds. The bonding agent was applied, adding the fresh resin onto the corresponding postpolymerized composite (Herculite/Herculite and Heliomolar/Heliomolar) by means of the silicon mold. Once the test specimens of each control and study group were prepared, they were stored dry at room temperature for 24 hours. For shear bond testing, the specimens were placed on fixing blocks (Fig. 4). The resulting block was placed in the test apparatus (Ibertest, Electrotest 500, Barcelona, Spain) with the use of a load cell of 500 dN. Shear bond strength was calculated by dividing the strength required to break the sample by the surface area of adhesion. The area of fracture of the samples was examined with a ×10 optic microscope to determine the primary fracture type (adhesive, cohesive, or mixed). For each study group, an additional sample was prepared for examination with an SEM (C Zeiss DSM 950).
Statistical analysis For each composite material, a 1-way analysis of variance (ANOVA) was performed. The ANOVA test showed that the effect of surface treatment on shear bond strength was significant for both materials. Consequently, the mean strengths of the experimental 483
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
Table III. Shear bond strengths of Heliomolar Radiopaque specimens
Fig. 4. Assembled specimen block before shear test.
Group
X ± SD (MPa)
Control 16 (air abrasion + Prime & Bond 2.0) 11 (air abrasion + Heliobond) 14 (air abrasion + Special Bond II) 18 (acetone + Prime & Bond 2.0) 13 (Special Bond II) 17 (hydrofluoric acid + Prime & Bond 2.0) 15 (phosphoric acid + Prime & Bond 2.0) 12 (hydrofluoric acid + Heliobond) 10 (phosphoric acid + Heliobond)
16.54 14.13 13.29 12.30 10.96 10.27 9.40 5.96 5.81 0.58
± ± ± ± ± ± ± ± ± ±
1.59 2.02 2.53 1.68 2.18 2.10 1.47 0.65 1.47 0.73
X = Mean. Values joined by vertical lines were not significantly different (P<.05).
Table II. Shear bond strengths of Herculite XRV specimens Group
Control 7 (air abrasion + Prime & Bond 2.0) 9 (acetone + Prime & Bond 2.0) 4 (Special Bond II) 2 (air abrasion + Heliobond) 5 (air abrasion + Special Bond II) 8 (hydrofluoric acid + Prime & Bond 2.0) 3 (hydrofluoric acid + Heliobond) 6 (phosphoric acid + Prime & Bond 2.0) 1 (phosphoric acid + Heliobond)
X ± SD (MPa)
17.76 14.65 13.07 12.59 11.85 10.74 10.91 8.40 5.37 0.10
± ± ± ± ± ± ± ± ± ±
1.59 4.90 1.32 1.32 0.80 4.64 1.14 3.04 0.62 0.22
X = Mean. Values joined by vertical lines were not significantly different (P<.05).
groups were compared with the use of the StudentNeuwman-Keuls test. To compare the shear bond strengths of Herculite and Heliomolar, the Student t test (with corrected df when variances were significantly different [Levene test, with P<.10]) was used within each group.
RESULTS The results of the shear bond test for Herculite and Heliomolar and the corresponding control groups are shown in Tables II and III, respectively. The Herculite specimens treated with air abrasion plus Prime & Bond 2.0 adhesive (group 7) showed statistically similar shear bond strengths to the control groups and were stronger than groups 5 (air abrasion, Special Bond II), 8 (hydrofluoric acid, Prime & Bond 2.0), 3 (hydrofluoric acid, Heliobond), 6 (phosphoric acid, Prime & Bond 2.0), and 1 (phosphoric acid, Heliobond). There were no significant differences between groups 7 (air abrasion, Prime & Bond 2.0), 9 (acetone, Prime & Bond 2.0), 4 (phosphoric acid, Special Bond II), or 2 (air abrasion, Heliobond). For Heliomolar, none of the surface treatments achieved statistically similar strengths to those of the 484
control samples. The air-abraded Heliomolar samples (groups 16, 11, and 14) presented higher bond strengths than those obtained in the remaining groups, except for those that received Special Bond II after air abrasion, for which the results were statistically similar to those obtained with acetone plus Prime & Bond 2.0 (group 18). There were no significant differences between groups 18 (acetone, Prime & Bond 2.0), 13 (Special Bond II), or 17 (hydrofluoric acid, Prime & Bond 2.0), which were all stronger than the groups treated with phosphoric/hydrofluoric acid plus Heliobond. The comparative results of Herculite and Heliomolar (Table IV) reveal that the type of composite influenced the results in 4 of the 9 protocols studied. When Heliomolar specimens were treated with hydrofluoric acid, the bond strength was significantly lower than that achieved in the same Herculite groups (group 3 vs 12 and group 8 vs 17). The bond strength of specimens treated with Special Bond II (group 4 vs 13) or acetone and Prime & Bond 2.0 (group 9 vs 18) was also lower in the Heliomolar groups. The optic microscope study of the specimens showed different fracture patterns (Table V). In the Heliomar group treated with phosphoric acid and Heliobond (group 10), 70% of the fractures were adhesive, whereas in the similarly treated Herculite group (group 1), 80% of the specimens exhibited bond failure during storage or transport or when they were mounted in the test apparatus. Among the specimens treated with air abrasion plus Special Bond II, 100% of fractured specimens in the Heliomolar group (group 14) showed adhesive failure, whereas 20% of the fractures in the Herculite group (group 5) were cohesive and within the postpolymerized composite. In the treatment group with the best results for both composites (groups 7 and 16), the fracture patterns changed: The fractures were mixed, with a predominance of cohesive fractures (Fig. 5). In the VOLUME 86 NUMBER 5
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
Table IV. Shear bond strength: Comparisons between Herculite and Heliomolar Shear bond strength (MPa) X ± SD Group
Control 1 vs 10 2 vs 11 3 vs 12 4 vs 13 5 vs 14 6 vs 15 7 vs 16 8 vs 17 9 vs 18
Herculite
17.76 0.10 11.85 8.40 12.59 10.74 5.37 14.65 10.91 13.07
± ± ± ± ± ± ± ± ± ±
1.59 0.22 0.80 3.04 1.32 4.64 0.62 4.90 1.14 1.32
Heliomolar
16.54 0.58 13.29 5.81 10.27 12.30 5.96 14.13 9.40 10.96
± ± ± ± ± ± ± ± ± ±
Herculite vs Heliomolar
1.59 0.73 2.53 1.47 2.10 1.68 0.65 2.02 1.47 2.18
texp(18 df) = 1.72, P=.103 (NS) texp(10.54 df) = 1.97, P=.076 (NS) texp(10.58 df) = 1.71, P=.116 (NS) texp(13 df) = 2.43, P=.0303 texp(15.15 df) = 2.95, P=.009 texp(11.33 df) = 0.99, P=.341 (NS) texp(17.96 df) = 2.08, P=.052 (NS) texp(11.96 df) = 0.31, P=.763 (NS) texp(16.95 df) = 2.56, P=.0202 texp(14.83 df) = 2.61, P=.0197
X = Mean; NS = not significant.
remaining study groups, the fracture pattern was similar for the 2 materials and always affected the interface between the postpolymerized composite and the new composite (Fig. 6).
DISCUSSION In this study, the shear bond strengths of Herculite and Heliomolar composites were found to be similar. In contrast, a study by Swift et al8 found that the bond strength of Herculite was greater than that of a microfilled composite (Silux), perhaps because after the control specimens were prepared they were postpolymerized again. In a study by Ferracane and Condon,12 the microfilled composites benefitted more from postpolymerization than did the hybrids because the initial low resistance to fracture of the microfilled composites increased after the laboratory processing, reaching similar levels to those of more densely loaded composites. Moreover, Silux is a BIS-GMA–based composite with polymerized particles, whereas Heliomolar is a urethane dimethacrylate-based composite with agglomerate complexes. Several studies6,7,13 have reported the need to condition highly polymerized composites to increase their adhesive capacity. In this study, the global efficacy of the different combinations of surface treatment and bonding agent varied widely. The use of phosphoric acid provided clinically unacceptable bond strengths in both microhybrid and microfilled composites. Söderholm10 argued that the unfilled resin cannot chemically react without a coupling agent, so that the only possibility of chemical bonding with the composite is by bonding with the residual monomers in the substrate. The better results for Heliomolar versus Herculite can be explained by the higher proportion of resin in microfilled composites. Hydrofluoric acid treatment produced significantly lower bond strength than did air abrasion or primer applications, which agrees with findings by Crumpler et al14 and Brosh et al,15 although Brosh et al studied NOVEMBER 2001
Table V. Optic microscopy results Composite
Herculite
Group
1 2 3 4 5 6 7 8 9 Heliomolar 10 11 12 13 14 15 16 17 18
No bond
80% — — — — — — — — 30% — — — — — — — —
Adhesive
20% 100% 100% 100% 80% 100% 10% 100% 100% 70% 100% 100% 100% 100% 100% 20% 100% 100%
Cohesive
Mixed
— — — — 20% — 70% — — — — — — — — 50% — —
— — — — — — 20% — — — — — — — — 30% — —
No bond: Specimens exhibited bond failure before the shear test (during storage, transport, or manipulation). Adhesive: Fracture between the postpolymerized and new composite. Cohesive: Fracture within the postpolymerized composite. Mixed: Adhesive-cohesive fracture.
a different microhybrid composite (Pertac hybrid). In agreement with Swift et al,8 the effect of hydrofluoric acid was related to the percentage, size, and type of the inorganic filling. Herculite was a better candidate for the use of hydrofluoric acid than Heliomolar (Figs. 7 and 8). In fact, the manufacturer (Kerr) recommends air abrasion plus hydrofluoric etching plus silane for incrustations on Herculite XRV. This protocol has not been tested, but various studies16,17 advise against this regimen because the hydrofluoric acid reduces the bond strength achieved with air abrasion. Hydrofluoric acid acts by dissolving the glass particles of the filling, leaving gaps or pores that allow micromechanical retention by the bonding agent.8 Its application at 9.6% for 2 minutes may have been too aggressive, because there 485
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 5. Cohesive fracture within postpolymerized composite.
Fig. 6. Adhesive fracture between postpolymerized and new composite.
are reports8 that etchings of shorter duration can dissolve the filling and soften the resin. The risks involved in handling hydrofluoric acid and the poor outcomes obtained may make it less than ideal as a surface preparation agent, especially for the repair of old composites of unknown chemical composition. Air abrasion provided the best results in this study (Figs. 9 and 10), which is in line with findings by other authors.8,13,16,17 The relative efficacy of the air abrasion treatments differed between Heliomolar and Herculite composites. For Heliomolar, air abrasion was unquestionably the most efficacious treatment. In fact, although the manufacturer recommends the use of Special Bond II to condition the microfilled composites, its application after air abrasion not only failed 486
LUCENA-MARTÍN ET AL
to increase the bond strength but also slightly reduced it (although the reduction did not reach statistical significance). In contrast, DeSchepper et al18 reported improvements in bond strength with the use of Special Bond I and II after air abrasion. However, they added the new composite immediately after the postpolymerization of the material; it is likely that the greater efficacy with the primer was due to the persistence of a greater amount of residual methacrylate groups. For Herculite, this study found no significant differences between the bond strength achieved with air abrasion plus Prime & Bond and that with air abrasion plus Heliobond or with acetone plus Prime & Bond. However, there were differences in the pattern of Herculite fractures, with air abrasion plus Heliobond producing 100% adhesive fractures and air abrasion plus Special Bond II producing 20% cohesive fractures. This suggests that primer treatment tends to improve the bond strength achieved by air abrasion, although this effect cannot be verified because of the high standard deviation in this group. An examination of the interaction between surface treatment and bonding agent revealed the following: When the postpolymerized composite was treated with orthophosphoric acid, the use of Prime & Bond 2.0 significantly increased the bond strength compared with the use of Heliobond, regardless of the type of composite; this finding agrees with that of other authors.6,13 The explanation may lie in the different capacities of the bonding agents to wet the substrate (composite). The viscosity of the unfilled adhesive resin Heliobond is considerably greater than that of Prime & Bond 2.0. Thus, the capacity of the latter to wet the surface and penetrate the organic phase of the composite is greater. A first reading of the results for interaction between bonding agent and air abrasion indicated that for both composite types, the influence of air abrasion on bond strength was greater than that of the bonding agent type. Nevertheless, even among the air-abraded composites, the use of a multifunctional adhesive produced a better outcome than the adhesive resin. For Herculite, the combination of air abrasion and Prime & Bond 2.0 was the only treatment that provided bond strengths statistically similar to the control group. Cohesive or cohesive-adhesive fractures were most common in this treatment group (for both Herculite and Heliomolar). Overall, the outcomes confirm the efficacy of 1-component adhesives in the repair of aged composites. The application of acetone plus Prime & Bond 2.0 adhesive achieved similar bond strengths to those obtained with the use of Special Bond II. This result was expected because both applications contain the same solvent in their chemical composition. VOLUME 86 NUMBER 5
LUCENA-MARTÍN ET AL
As mentioned previously, the effect of hydrofluoric acid varied according to the composite to which it was applied. The effect of primer (Special Bond II/acetone) application also differed, producing weaker bond strengths in microfilled than in microhybrid composites. This result is paradoxical given the greater proportion of resin in the microfilled composites, but it agrees with findings by Imamura et al,16 who tested a different microfilled composite (Concept). No data could be found in the literature on the adequate bond strengths needed for a repair to survive in occlusal function, although composite repair bond strengths greater than 18 MPa have been reported to give clinically acceptable results.6,7 Swift et al17 found that microabrasion plus phosphinate adhesive (Bondlite) on 7-day-old composites achieved bond strengths of 45% to 70% of the cohesive strength of the composite. Other studies13,16 that used a similar protocol showed adhesion strengths of 20 to 25 MPa. A direct comparison of the numerical results obtained by different authors would prove confusing, because their respective experimental conditions were not standardized. In this study, the low results obtained for groups 1 and 10 may be attributed to the fact that these experimental groups reflected the reaction of the unfilled resin (Heliobond) with the residual unsaturated monomers in the composite substrate.6 With the exception of these groups, the results ranged between 30% and 82.5% of the cohesive strength of the composite for Herculite and 35% and 85% for Heliomolar; such results are consistent with those obtained by Swift et al.17 If a composite repair tends to fracture within the original composite (cohesive fracture), one can assume the selected protocol to be appropriate to bear the occlusal loads, provided that the initial composite chosen was appropriate for the restorative application.19 This experimental study design provided no data on the long-term stability of the adhesion achieved. In a study by Sau et al,20 there was a general increase in the repair shear bond strengths at 1 week and general deterioration at 4 weeks in most of the repaired materials. Further studies are required to address the effect of thermal cycling and long-term storage in a moist environment on repair shear bond strengths.
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 7. Herculite XRV specimen treated with hydrofluoric acid.
Fig. 8. Heliomolar Radiopaque specimen treated with hydrofluoric acid.
CONCLUSIONS Within the limitations of this study, the following conclusions were drawn: 1. Among the surface treatments tested, air abrasion produced the strongest bonds. 2. The composition of the composite did not modify the effect of air abrasion on bond strength. 3. The use of Prime & Bond 2.0 adhesive resulted in greater bond strengths than those produced with Heliobond. NOVEMBER 2001
Fig. 9. Herculite XRV specimen after treatment with air abrasion.
4. There was a synergic effect between mechanical surface treatments and dentin adhesives. The combination that produced bond strengths closest to the 487
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
8. 9. 10. 11.
12. 13. 14. 15.
Fig. 10. Heliomolar Radiopaque specimen after treatment with air abrasion.
16. 17. 18.
cohesive strength of the composites, for both microfilled and microhybrid composites, was air abrasion plus the application of a 1-component adhesive.
19. 20.
posite resin with a direct, visible-light-cured composite resin. Oper Dent 1993;18:187-94. Swift EJ Jr, Le Valley BD, Boyer DB. Evaluation of new methods for composite repair. Dent Mater 1992;8:362-5. Causton BE. Repair of abraded composite fillings. An in vitro study. Br Dent J 1975;139:286-8. Söderholm KJ. Flexure strength of repaired dental composites. Scand J Dent Res 1986;94:364-9. Lucena C, González MP, Ferrer CM, Robles V, Navajas JM. Study of the shear bond strength of five one-component adhesives under simulated pulpal pressure. Oper Dent 1999;24:73-80. Ferracane JL, Condon JR. Post-cure heat treatments for composites: properties and fractography. Dent Mater 1992;8:290-5. Kupiec KA, Barkmeier WW. Laboratory evaluation of surface treatments for composite repair. Oper Dent 1996;21:59-62. Crumpler DC, Bayne SC, Sockwell S, Brunson D, Roberson TM. Bonding to resurfaced posterior composites. Dent Mater 1989;5:417-24. Brosh T, Pilo R, Bichacho N, Blutstein R. Effects of combinations of surface treatments and bonding agents on the bond strength of repaired composites. J Prosthet Dent 1997;77:122-6. Imamura GM, Reinhardt JW, Boyer DB, Swift EJ Jr. Enhancement of resin bonding to heat-cured composite resin. Oper Dent 1996,21:249-56. Swift EJ Jr, Brodeur C, Cvitko E, Pires JA. Treatment of composite surfaces for indirect bonding. Dent Mater 1992;8:193-6. DeSchepper EJ, Tate WH, Powers JM. Bond strength of resin cements to microfilled composites. Am J Dent 1993;6:235-8. Bouschlicher MR, Reinhardt JW, Vargas MA. Surface treatment techniques for resin composite repair. Am J Dent 1997;10:279-83. Sau CW, Oh GS, Koh H, Chee CS, Lim CC. Shear bond strength of repaired composite resins using a hybrid composite resin. Oper Dent 1999;24:156-61.
REFERENCES 1. Mjör IA. Repair versus replacement of failed restorations. Int Dent J 1993;43:466-72. 2. Lloyd CH, Baigrie DA, Jeffrey IW. The tensile strength of composite repairs. J Dent 1980;8:171-7. 3. Boyer DB, Chan KC, Reinhardt JW. Build-up and repair of light-cured composites: bond strength. J Dent Res 1984;63:1241-4. 4. Azarbal P, Boyer DB, Chan KC. The effect of bonding agents on the interfacial bond strength of repaired composites. Dent Mater 1986;2:153-5. 5. Boyer DB, Chan KC, Torney DL. The strength of multilayer and repaired composite resin. J Prosthet Dent 1978;39:63-7. 6. Puckett AD, Holder R, O’Hara JW. Strength of posterior composite repairs using different composite/bonding agent combinations. Oper Dent 1991;16:136-40. 7. Turner CW, Meiers JC. Repair of an aged, contaminated indirect com-
488
Reprint requests to: DR C. LUCENA-MARTÍN C, MURALLA, Nο 1, 3A 18600 MOTRIL GRANADA, SPAIN FAX: (34)958-243-795 E-MAIL: [email protected] Copyright © 2001 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/2001/$35.00 + 0. 10/1/116775
doi:10.1067/mpr.2001.116775
VOLUME 86 NUMBER 5
CLINICAL IMPLICATIONS Among the protocols tested, surface treatment with air abrasion and a 1-component adhesive most effectively enhanced the repair of aged and heat-polymerized microhybrid and microfilled composites.
T
he rising patient demand for esthetic dental restorations coupled with concerns about the toxicity of amalgam has markedly increased the use of composites. Despite some recent improvements, their technique sensitivity leads to numerous failures in the clinical setting, especially when they are used in posterior teeth. The repair of composites that exhibit small fractures, staining, or wear may be a viable and less costly alternative to their complete replacement and would cause less pulpal trauma.1 The bond strength between increments of composite is equal to the cohesive strength of the material.2,3 However, if the composite has been contaminated, polished, processed in a laboratory (indirect composite restorations), or aged, the adhesion to a new composite is reduced to 25% to 80% of the original cohesive strength.4-7 Various methods have been reported to improve the reactivity of highly converted composites; these methaAssistant
Professor, Department of Pathology and Dental Therapy. Professor, Department of Pathology and Dental Therapy. cProfessor, Department of Pathology and Dental Therapy. bAssociate
NOVEMBER 2001
ods include acid etching,8 air abrasion,5,9 and the use of solvents and silanes.10 There is no consensus on the results obtained with the different procedures. Several studies have supported the efficacy of multifunctional 1-component adhesives and their bond to enamel and dentin.11 However, their efficacy for composite repairs has not been adequately studied. This study was designed to evaluate the efficacy of different mechanical and chemical procedures used to improve the bond strength of highly polymerized microhybrid and microfilled composites.
MATERIAL AND METHODS The composites chosen for this study were a hybrid composite containing a BIS-GMA matrix resin (Herculite XRV; Kerr Manufacturing, Romulus, Mich.) and a microfilled composite based primarily on urethane dimethacrylate (Heliomolar Radiopaque; Vivadent, Schaan, Liechtenstein). Ten specimens of Herculite XRV and Heliomolar Radiopaque composite were prepared as controls. Twenty aluminum molds consisting of a disk 20 mm in external diameter and 3 mm in depth with a central THE JOURNAL OF PROSTHETIC DENTISTRY 481
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 1. Aluminum mold.
LUCENA-MARTÍN ET AL
cavity 5 and 8 mm in diameter (top and bottom base, respectively) were used (Fig. 1). The mold was placed on a glass slide, and a single increment of composite was condensed in each by means of a plastic instrument. The resin was immediately light-polymerized for 40 seconds, with the tip of the light-polymerizing unit (Optilux 401; Demetron Research Corp, Danbury, Conn.) placed in contact with the surface of the mold (Fig. 2). Light-polymerization was repeated for 40 seconds from underneath the glass slide. A similar polymerization area was achieved for all specimens, as each was maintained the same distance from the light source. A silicon disk (20 mm in external diameter and 4 mm in depth, with a central cavity 5 mm in diameter) was placed on the aluminum mold, and composite was condensed into it in 2 increments: The first increment filled two thirds of the cylinder height, and the second increment filled it to the top. Each increment was light-polymerized for 40 seconds. The resulting specimens were submitted to a secondary postpolymerization cycle (Ivomat 500; Ivoclar AG FL-9494, Schaan, Liechtenstein) with a combined application of heat (120°C) and pressure (6 atmospheres) for 10 minutes, within a maximum of 6 hours.
Study groups
Fig. 2. Light-polymerization of composite.
Fig. 3. Slide placed over fresh composite.
482
One hundred eighty aluminum molds were divided into 2 groups (n = 90 each), 1 group for Herculite XRV and 1 group for Heliomolar Radiopaque composite. Each mold was placed on a glass slide. Resin was placed within the mold in a single increment, and the first slide then was covered with another glass slide (Fig. 3). The composite was polymerized through the top slide to achieve a glossy, smooth surface. Because the total height of the mold was 3 mm, it was lightpolymerized for an additional 40 seconds through the bottom slide. Before the study and before each specimen group was polymerized, the polymerizing light output was tested with a Cure-Rite light intensity meter (Efos Inc, Mississagwa, Ontario, Canada). This ensured that the light intensity was not less than 300 W/m2. After secondary polymerization (120°C per 6 atmospheres per 10 minutes), the specimens were cooled for 10 minutes and then placed in an ultrasound bath for 2 minutes to remove any surface debris left behind by the postpolymerization treatment. The specimens were stored in distilled water for 4 weeks at room temperature. The specimens of each group were randomly assigned to 1 of 9 study subgroups for different combinations of surface treatment and bonding agents (Table I). The treatment protocols involved the use of air abrasion, 37% orthophosphoric acid, 9.6% hydrofluoric acid, Special Bond II, 99% acetone, Heliobond, and/or Prime & Bond 2.0. VOLUME 86 NUMBER 5
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
Table I. Treatment applied to composite Composite
Herculite
Heliomolar
Group
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Air abrasion
H3PO4
+
+ +
HF
Acetone
SB II
+ +
+
+ + + + + +
+ + + + +
+ + +
+ + + + + +
P&B 2.0
+ + +
+ + + + + + + + +
Heliobond
+ +
+ +
+ + + +
H3PO4 = 37% phosphoric acid; HF = 9.6% hydrofluoric acid; SB II = Special Bond II; P&B 2.0 = Prime & Bond 2.0.
Air abrasion: A flow of 50 µm aluminum oxide particles was used at a pressure of 60 psi (Microetcher; Danville Engineering, Danville, Calif.). The orifice of the instrument was placed 1 cm from the composite surface until the glossy appearance of the composite turned to a matte finish. 37% orthophosphoric acid: The acid (Email Preparator GS; Ivoclar) was applied for 1 minute. The surface of the resin then was washed for 45 seconds and dried with compressed air. 9.6% hydrofluoric acid: The acid (Bite Perf, Malaga, Spain) was applied for 2 minutes. The specimen then was carefully washed and dried with compressed air. Special Bond II: The surface of the resin first was treated with orthophosphoric acid. After a thin layer of Special Bond II (Vivadent) was applied according to the manufacturer’s instructions, the surface was dried for 20 seconds and light-polymerized for 40 seconds. An adhesive resin was not used because Special Bond II already contains an adhesive. 99% acetone: A brush was dipped several times into the acetone solution (Panreac; Monfplet & Esteban SA, Barcelona, Spain), which was used to coat the surface of the composite for 1 minute. The surface then was lightly dried to evaporate the remaining solvent. Heliobond: One increment of Heliobond (Vivadent) was applied, thinned with an air syringe, and then light-polymerized for 10 seconds. Prime & Bond 2.0: The adhesive (Dentsply/DeTrey, Konstanz, Germany) was applied according to the manufacturer’s instructions. Regardless of the type of surface treatment, the postpolymerized composite was treated with 37% phosphoric acid for 15 seconds. After NOVEMBER 2001
the washing and drying steps, a generous layer of Prime & Bond 2.0 was applied, ensuring that the surface of the resin remained wet for 20 seconds. The remaining solvent was removed with compressed air, and the resin was light-polymerized for 10 seconds. A second increment of adhesive was applied, the excess solvent was removed, and the resin was again lightpolymerized for 10 seconds. The bonding agent was applied, adding the fresh resin onto the corresponding postpolymerized composite (Herculite/Herculite and Heliomolar/Heliomolar) by means of the silicon mold. Once the test specimens of each control and study group were prepared, they were stored dry at room temperature for 24 hours. For shear bond testing, the specimens were placed on fixing blocks (Fig. 4). The resulting block was placed in the test apparatus (Ibertest, Electrotest 500, Barcelona, Spain) with the use of a load cell of 500 dN. Shear bond strength was calculated by dividing the strength required to break the sample by the surface area of adhesion. The area of fracture of the samples was examined with a ×10 optic microscope to determine the primary fracture type (adhesive, cohesive, or mixed). For each study group, an additional sample was prepared for examination with an SEM (C Zeiss DSM 950).
Statistical analysis For each composite material, a 1-way analysis of variance (ANOVA) was performed. The ANOVA test showed that the effect of surface treatment on shear bond strength was significant for both materials. Consequently, the mean strengths of the experimental 483
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
Table III. Shear bond strengths of Heliomolar Radiopaque specimens
Fig. 4. Assembled specimen block before shear test.
Group
X ± SD (MPa)
Control 16 (air abrasion + Prime & Bond 2.0) 11 (air abrasion + Heliobond) 14 (air abrasion + Special Bond II) 18 (acetone + Prime & Bond 2.0) 13 (Special Bond II) 17 (hydrofluoric acid + Prime & Bond 2.0) 15 (phosphoric acid + Prime & Bond 2.0) 12 (hydrofluoric acid + Heliobond) 10 (phosphoric acid + Heliobond)
16.54 14.13 13.29 12.30 10.96 10.27 9.40 5.96 5.81 0.58
± ± ± ± ± ± ± ± ± ±
1.59 2.02 2.53 1.68 2.18 2.10 1.47 0.65 1.47 0.73
X = Mean. Values joined by vertical lines were not significantly different (P<.05).
Table II. Shear bond strengths of Herculite XRV specimens Group
Control 7 (air abrasion + Prime & Bond 2.0) 9 (acetone + Prime & Bond 2.0) 4 (Special Bond II) 2 (air abrasion + Heliobond) 5 (air abrasion + Special Bond II) 8 (hydrofluoric acid + Prime & Bond 2.0) 3 (hydrofluoric acid + Heliobond) 6 (phosphoric acid + Prime & Bond 2.0) 1 (phosphoric acid + Heliobond)
X ± SD (MPa)
17.76 14.65 13.07 12.59 11.85 10.74 10.91 8.40 5.37 0.10
± ± ± ± ± ± ± ± ± ±
1.59 4.90 1.32 1.32 0.80 4.64 1.14 3.04 0.62 0.22
X = Mean. Values joined by vertical lines were not significantly different (P<.05).
groups were compared with the use of the StudentNeuwman-Keuls test. To compare the shear bond strengths of Herculite and Heliomolar, the Student t test (with corrected df when variances were significantly different [Levene test, with P<.10]) was used within each group.
RESULTS The results of the shear bond test for Herculite and Heliomolar and the corresponding control groups are shown in Tables II and III, respectively. The Herculite specimens treated with air abrasion plus Prime & Bond 2.0 adhesive (group 7) showed statistically similar shear bond strengths to the control groups and were stronger than groups 5 (air abrasion, Special Bond II), 8 (hydrofluoric acid, Prime & Bond 2.0), 3 (hydrofluoric acid, Heliobond), 6 (phosphoric acid, Prime & Bond 2.0), and 1 (phosphoric acid, Heliobond). There were no significant differences between groups 7 (air abrasion, Prime & Bond 2.0), 9 (acetone, Prime & Bond 2.0), 4 (phosphoric acid, Special Bond II), or 2 (air abrasion, Heliobond). For Heliomolar, none of the surface treatments achieved statistically similar strengths to those of the 484
control samples. The air-abraded Heliomolar samples (groups 16, 11, and 14) presented higher bond strengths than those obtained in the remaining groups, except for those that received Special Bond II after air abrasion, for which the results were statistically similar to those obtained with acetone plus Prime & Bond 2.0 (group 18). There were no significant differences between groups 18 (acetone, Prime & Bond 2.0), 13 (Special Bond II), or 17 (hydrofluoric acid, Prime & Bond 2.0), which were all stronger than the groups treated with phosphoric/hydrofluoric acid plus Heliobond. The comparative results of Herculite and Heliomolar (Table IV) reveal that the type of composite influenced the results in 4 of the 9 protocols studied. When Heliomolar specimens were treated with hydrofluoric acid, the bond strength was significantly lower than that achieved in the same Herculite groups (group 3 vs 12 and group 8 vs 17). The bond strength of specimens treated with Special Bond II (group 4 vs 13) or acetone and Prime & Bond 2.0 (group 9 vs 18) was also lower in the Heliomolar groups. The optic microscope study of the specimens showed different fracture patterns (Table V). In the Heliomar group treated with phosphoric acid and Heliobond (group 10), 70% of the fractures were adhesive, whereas in the similarly treated Herculite group (group 1), 80% of the specimens exhibited bond failure during storage or transport or when they were mounted in the test apparatus. Among the specimens treated with air abrasion plus Special Bond II, 100% of fractured specimens in the Heliomolar group (group 14) showed adhesive failure, whereas 20% of the fractures in the Herculite group (group 5) were cohesive and within the postpolymerized composite. In the treatment group with the best results for both composites (groups 7 and 16), the fracture patterns changed: The fractures were mixed, with a predominance of cohesive fractures (Fig. 5). In the VOLUME 86 NUMBER 5
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
Table IV. Shear bond strength: Comparisons between Herculite and Heliomolar Shear bond strength (MPa) X ± SD Group
Control 1 vs 10 2 vs 11 3 vs 12 4 vs 13 5 vs 14 6 vs 15 7 vs 16 8 vs 17 9 vs 18
Herculite
17.76 0.10 11.85 8.40 12.59 10.74 5.37 14.65 10.91 13.07
± ± ± ± ± ± ± ± ± ±
1.59 0.22 0.80 3.04 1.32 4.64 0.62 4.90 1.14 1.32
Heliomolar
16.54 0.58 13.29 5.81 10.27 12.30 5.96 14.13 9.40 10.96
± ± ± ± ± ± ± ± ± ±
Herculite vs Heliomolar
1.59 0.73 2.53 1.47 2.10 1.68 0.65 2.02 1.47 2.18
texp(18 df) = 1.72, P=.103 (NS) texp(10.54 df) = 1.97, P=.076 (NS) texp(10.58 df) = 1.71, P=.116 (NS) texp(13 df) = 2.43, P=.0303 texp(15.15 df) = 2.95, P=.009 texp(11.33 df) = 0.99, P=.341 (NS) texp(17.96 df) = 2.08, P=.052 (NS) texp(11.96 df) = 0.31, P=.763 (NS) texp(16.95 df) = 2.56, P=.0202 texp(14.83 df) = 2.61, P=.0197
X = Mean; NS = not significant.
remaining study groups, the fracture pattern was similar for the 2 materials and always affected the interface between the postpolymerized composite and the new composite (Fig. 6).
DISCUSSION In this study, the shear bond strengths of Herculite and Heliomolar composites were found to be similar. In contrast, a study by Swift et al8 found that the bond strength of Herculite was greater than that of a microfilled composite (Silux), perhaps because after the control specimens were prepared they were postpolymerized again. In a study by Ferracane and Condon,12 the microfilled composites benefitted more from postpolymerization than did the hybrids because the initial low resistance to fracture of the microfilled composites increased after the laboratory processing, reaching similar levels to those of more densely loaded composites. Moreover, Silux is a BIS-GMA–based composite with polymerized particles, whereas Heliomolar is a urethane dimethacrylate-based composite with agglomerate complexes. Several studies6,7,13 have reported the need to condition highly polymerized composites to increase their adhesive capacity. In this study, the global efficacy of the different combinations of surface treatment and bonding agent varied widely. The use of phosphoric acid provided clinically unacceptable bond strengths in both microhybrid and microfilled composites. Söderholm10 argued that the unfilled resin cannot chemically react without a coupling agent, so that the only possibility of chemical bonding with the composite is by bonding with the residual monomers in the substrate. The better results for Heliomolar versus Herculite can be explained by the higher proportion of resin in microfilled composites. Hydrofluoric acid treatment produced significantly lower bond strength than did air abrasion or primer applications, which agrees with findings by Crumpler et al14 and Brosh et al,15 although Brosh et al studied NOVEMBER 2001
Table V. Optic microscopy results Composite
Herculite
Group
1 2 3 4 5 6 7 8 9 Heliomolar 10 11 12 13 14 15 16 17 18
No bond
80% — — — — — — — — 30% — — — — — — — —
Adhesive
20% 100% 100% 100% 80% 100% 10% 100% 100% 70% 100% 100% 100% 100% 100% 20% 100% 100%
Cohesive
Mixed
— — — — 20% — 70% — — — — — — — — 50% — —
— — — — — — 20% — — — — — — — — 30% — —
No bond: Specimens exhibited bond failure before the shear test (during storage, transport, or manipulation). Adhesive: Fracture between the postpolymerized and new composite. Cohesive: Fracture within the postpolymerized composite. Mixed: Adhesive-cohesive fracture.
a different microhybrid composite (Pertac hybrid). In agreement with Swift et al,8 the effect of hydrofluoric acid was related to the percentage, size, and type of the inorganic filling. Herculite was a better candidate for the use of hydrofluoric acid than Heliomolar (Figs. 7 and 8). In fact, the manufacturer (Kerr) recommends air abrasion plus hydrofluoric etching plus silane for incrustations on Herculite XRV. This protocol has not been tested, but various studies16,17 advise against this regimen because the hydrofluoric acid reduces the bond strength achieved with air abrasion. Hydrofluoric acid acts by dissolving the glass particles of the filling, leaving gaps or pores that allow micromechanical retention by the bonding agent.8 Its application at 9.6% for 2 minutes may have been too aggressive, because there 485
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 5. Cohesive fracture within postpolymerized composite.
Fig. 6. Adhesive fracture between postpolymerized and new composite.
are reports8 that etchings of shorter duration can dissolve the filling and soften the resin. The risks involved in handling hydrofluoric acid and the poor outcomes obtained may make it less than ideal as a surface preparation agent, especially for the repair of old composites of unknown chemical composition. Air abrasion provided the best results in this study (Figs. 9 and 10), which is in line with findings by other authors.8,13,16,17 The relative efficacy of the air abrasion treatments differed between Heliomolar and Herculite composites. For Heliomolar, air abrasion was unquestionably the most efficacious treatment. In fact, although the manufacturer recommends the use of Special Bond II to condition the microfilled composites, its application after air abrasion not only failed 486
LUCENA-MARTÍN ET AL
to increase the bond strength but also slightly reduced it (although the reduction did not reach statistical significance). In contrast, DeSchepper et al18 reported improvements in bond strength with the use of Special Bond I and II after air abrasion. However, they added the new composite immediately after the postpolymerization of the material; it is likely that the greater efficacy with the primer was due to the persistence of a greater amount of residual methacrylate groups. For Herculite, this study found no significant differences between the bond strength achieved with air abrasion plus Prime & Bond and that with air abrasion plus Heliobond or with acetone plus Prime & Bond. However, there were differences in the pattern of Herculite fractures, with air abrasion plus Heliobond producing 100% adhesive fractures and air abrasion plus Special Bond II producing 20% cohesive fractures. This suggests that primer treatment tends to improve the bond strength achieved by air abrasion, although this effect cannot be verified because of the high standard deviation in this group. An examination of the interaction between surface treatment and bonding agent revealed the following: When the postpolymerized composite was treated with orthophosphoric acid, the use of Prime & Bond 2.0 significantly increased the bond strength compared with the use of Heliobond, regardless of the type of composite; this finding agrees with that of other authors.6,13 The explanation may lie in the different capacities of the bonding agents to wet the substrate (composite). The viscosity of the unfilled adhesive resin Heliobond is considerably greater than that of Prime & Bond 2.0. Thus, the capacity of the latter to wet the surface and penetrate the organic phase of the composite is greater. A first reading of the results for interaction between bonding agent and air abrasion indicated that for both composite types, the influence of air abrasion on bond strength was greater than that of the bonding agent type. Nevertheless, even among the air-abraded composites, the use of a multifunctional adhesive produced a better outcome than the adhesive resin. For Herculite, the combination of air abrasion and Prime & Bond 2.0 was the only treatment that provided bond strengths statistically similar to the control group. Cohesive or cohesive-adhesive fractures were most common in this treatment group (for both Herculite and Heliomolar). Overall, the outcomes confirm the efficacy of 1-component adhesives in the repair of aged composites. The application of acetone plus Prime & Bond 2.0 adhesive achieved similar bond strengths to those obtained with the use of Special Bond II. This result was expected because both applications contain the same solvent in their chemical composition. VOLUME 86 NUMBER 5
LUCENA-MARTÍN ET AL
As mentioned previously, the effect of hydrofluoric acid varied according to the composite to which it was applied. The effect of primer (Special Bond II/acetone) application also differed, producing weaker bond strengths in microfilled than in microhybrid composites. This result is paradoxical given the greater proportion of resin in the microfilled composites, but it agrees with findings by Imamura et al,16 who tested a different microfilled composite (Concept). No data could be found in the literature on the adequate bond strengths needed for a repair to survive in occlusal function, although composite repair bond strengths greater than 18 MPa have been reported to give clinically acceptable results.6,7 Swift et al17 found that microabrasion plus phosphinate adhesive (Bondlite) on 7-day-old composites achieved bond strengths of 45% to 70% of the cohesive strength of the composite. Other studies13,16 that used a similar protocol showed adhesion strengths of 20 to 25 MPa. A direct comparison of the numerical results obtained by different authors would prove confusing, because their respective experimental conditions were not standardized. In this study, the low results obtained for groups 1 and 10 may be attributed to the fact that these experimental groups reflected the reaction of the unfilled resin (Heliobond) with the residual unsaturated monomers in the composite substrate.6 With the exception of these groups, the results ranged between 30% and 82.5% of the cohesive strength of the composite for Herculite and 35% and 85% for Heliomolar; such results are consistent with those obtained by Swift et al.17 If a composite repair tends to fracture within the original composite (cohesive fracture), one can assume the selected protocol to be appropriate to bear the occlusal loads, provided that the initial composite chosen was appropriate for the restorative application.19 This experimental study design provided no data on the long-term stability of the adhesion achieved. In a study by Sau et al,20 there was a general increase in the repair shear bond strengths at 1 week and general deterioration at 4 weeks in most of the repaired materials. Further studies are required to address the effect of thermal cycling and long-term storage in a moist environment on repair shear bond strengths.
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 7. Herculite XRV specimen treated with hydrofluoric acid.
Fig. 8. Heliomolar Radiopaque specimen treated with hydrofluoric acid.
CONCLUSIONS Within the limitations of this study, the following conclusions were drawn: 1. Among the surface treatments tested, air abrasion produced the strongest bonds. 2. The composition of the composite did not modify the effect of air abrasion on bond strength. 3. The use of Prime & Bond 2.0 adhesive resulted in greater bond strengths than those produced with Heliobond. NOVEMBER 2001
Fig. 9. Herculite XRV specimen after treatment with air abrasion.
4. There was a synergic effect between mechanical surface treatments and dentin adhesives. The combination that produced bond strengths closest to the 487
THE JOURNAL OF PROSTHETIC DENTISTRY
LUCENA-MARTÍN ET AL
8. 9. 10. 11.
12. 13. 14. 15.
Fig. 10. Heliomolar Radiopaque specimen after treatment with air abrasion.
16. 17. 18.
cohesive strength of the composites, for both microfilled and microhybrid composites, was air abrasion plus the application of a 1-component adhesive.
19. 20.
posite resin with a direct, visible-light-cured composite resin. Oper Dent 1993;18:187-94. Swift EJ Jr, Le Valley BD, Boyer DB. Evaluation of new methods for composite repair. Dent Mater 1992;8:362-5. Causton BE. Repair of abraded composite fillings. An in vitro study. Br Dent J 1975;139:286-8. Söderholm KJ. Flexure strength of repaired dental composites. Scand J Dent Res 1986;94:364-9. Lucena C, González MP, Ferrer CM, Robles V, Navajas JM. Study of the shear bond strength of five one-component adhesives under simulated pulpal pressure. Oper Dent 1999;24:73-80. Ferracane JL, Condon JR. Post-cure heat treatments for composites: properties and fractography. Dent Mater 1992;8:290-5. Kupiec KA, Barkmeier WW. Laboratory evaluation of surface treatments for composite repair. Oper Dent 1996;21:59-62. Crumpler DC, Bayne SC, Sockwell S, Brunson D, Roberson TM. Bonding to resurfaced posterior composites. Dent Mater 1989;5:417-24. Brosh T, Pilo R, Bichacho N, Blutstein R. Effects of combinations of surface treatments and bonding agents on the bond strength of repaired composites. J Prosthet Dent 1997;77:122-6. Imamura GM, Reinhardt JW, Boyer DB, Swift EJ Jr. Enhancement of resin bonding to heat-cured composite resin. Oper Dent 1996,21:249-56. Swift EJ Jr, Brodeur C, Cvitko E, Pires JA. Treatment of composite surfaces for indirect bonding. Dent Mater 1992;8:193-6. DeSchepper EJ, Tate WH, Powers JM. Bond strength of resin cements to microfilled composites. Am J Dent 1993;6:235-8. Bouschlicher MR, Reinhardt JW, Vargas MA. Surface treatment techniques for resin composite repair. Am J Dent 1997;10:279-83. Sau CW, Oh GS, Koh H, Chee CS, Lim CC. Shear bond strength of repaired composite resins using a hybrid composite resin. Oper Dent 1999;24:156-61.
REFERENCES 1. Mjör IA. Repair versus replacement of failed restorations. Int Dent J 1993;43:466-72. 2. Lloyd CH, Baigrie DA, Jeffrey IW. The tensile strength of composite repairs. J Dent 1980;8:171-7. 3. Boyer DB, Chan KC, Reinhardt JW. Build-up and repair of light-cured composites: bond strength. J Dent Res 1984;63:1241-4. 4. Azarbal P, Boyer DB, Chan KC. The effect of bonding agents on the interfacial bond strength of repaired composites. Dent Mater 1986;2:153-5. 5. Boyer DB, Chan KC, Torney DL. The strength of multilayer and repaired composite resin. J Prosthet Dent 1978;39:63-7. 6. Puckett AD, Holder R, O’Hara JW. Strength of posterior composite repairs using different composite/bonding agent combinations. Oper Dent 1991;16:136-40. 7. Turner CW, Meiers JC. Repair of an aged, contaminated indirect com-
488
Reprint requests to: DR C. LUCENA-MARTÍN C, MURALLA, Nο 1, 3A 18600 MOTRIL GRANADA, SPAIN FAX: (34)958-243-795 E-MAIL: [email protected] Copyright © 2001 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/2001/$35.00 + 0. 10/1/116775
doi:10.1067/mpr.2001.116775
VOLUME 86 NUMBER 5