Two-Year Interfacial Bond Durability and Nanoleakage of Repaired Silorane-Based Resin Composite Open Access

The concept of minimally invasive dentistry opened the way for conservative tooth restorations as well as for repairing defective preexisting restorations rather than completely replacement them.¹  Researchers have shown that the repair of tooth-colored restorations is a conservative, time-saving, cost-reducing, effective method in which the intact part of the restoration can be maintained and unnecessary removal of dental hard tissues and pulp irritation is avoided.^1–3 

Currently, different tooth-colored resin composites are available, meaning that composite monomers other than dimethacrylates are being used. In 2005, a silorane-based composite was introduced. Because of its modified matrix consisting of siloxane and oxirane components, silorane-based composite exhibits a reduced shrinkage of approximately 1% by volume due to ring-opening cationic polymerization.⁴  Because of its recent introduction, little is known about the repairing ability of this category of resin composites.

Different repair approaches of silorane composites were recently investigated.^5–11  The effect of intermediate adhesive agent/repair composite,^5,8  the impact of surface preparation,^7,9,11  or both^6,10  on silorane-based composite repair was tested. In these studies, some researchers tested the ability of methacrylate-based adhesive/composite to repair silorane restorations.^5,8–10  One study proved that fresh silorane composite can be equally repaired with the same material or with methacrylate-based composite.⁹  Others^5,8,10  recommended the application of a silane-coupling agent to enhance the repair bond strength of methacrylate-based composites to aged silorane-based resin composite. However, separate application of a silane primer and a dentin adhesive can result in a thick, multiphase interfacial layer, which may introduce defects in each working step.¹²  None of these studies evaluated nanoleakage with silver nitrate uptake. However, such analysis would provide good spatial resolution of submicron defects in such an interfacial layer.¹³ 

Although promising results were obtained from these studies, the results were about short-term bonding, which would not predict long-term clinical performance. Interfacial bond of the repaired composite may be impaired by mechanical, thermal, and chemical stresses in the intraoral environment,¹⁴  mainly as a consequence of the limited hydrolytic stability of intermediate agents (ie, silanes and/or adhesives).¹⁵  Thus, long-term water storage of specimens was recommended as a validated method for assessment of bond degradation.¹⁶ 

On reviewing the literature, no published data could be retrieved about the long-term repair bond durability of silorane-based resin composite. Thus, it would be of benefit to investigate the repair bond durability of silorane-based resin composite by means of microtensile bond strength (μTBS) and nanoleakage, when repaired with the same material or with methacrylate-based adhesive/composite with or without using silane primer as a repair bonding promoter. The null hypotheses tested were 1) the use of silane primer and the type of intermediate adhesive agents/repair composites do not affect the interfacial μTBS of repaired silorane-based resin composites; 2) there is no difference between silorane interfacial repair bond strength after short-term (24 hours) and after long-term (two years) storage in artificial saliva; 3) silane primer, different intermediate adhesive agent/repair composite, as well as different storage periods have no influence on interfacial nanoleakage of repaired silorane-based resin composites.

The intermediate adhesive agents' brand names, manufacturers, chemical compositions, and batch numbers, as well as their steps of application, are listed in Table 1. Repair resin composite materials' brand names, manufacturers, batch numbers, and chemical compositions are listed in Table 2.

A circular split Teflon mold (substrate mold) 30 mm in diameter and 4 mm in height was specially constructed. It had a central hole of 4 mm diameter in which the substrate resin composite specimen was made. A second mold (repair mold) enclosed three split Teflon discs. The base disc (20 mm external diameter and 3.5 mm height) had a hole of 4 mm internal diameter and was used to hold the substrate specimen while applying the adhesive system. The other two discs were used on top of the base disc to build up the repair material. Both discs had external and hole diameters similar to the base disc. However, one was 2 mm in height and the other was 1.5 mm in height.⁹  The repair mold is presented in Figure 1.

A total of 48 Filtek P90 substrate specimens (4 mm diameter and 4 mm height) were made. The substrate mold was placed on a celluloid strip matrix (Dental Technologies Inc, Lincolnwood, IL, USA) and a glass slab. Filtek P90 shade A3 was packed in the mold in one increment of 4-mm thickness. The top surface of the substrate specimen was covered with a celluloid strip to prevent the oxygen inhibited layer. It was pressed using a glass slab to remove excess composite material before curing and to gain smooth surface of the specimen. The glass slab was removed, and each specimen was light cured using an LED light-curing unit (Blue Phase C5, Ivoclar Vivadent, Schaan, Lechtenstein) from its top and bottom for 40 seconds each. The specimen was then removed from the mold and cured from both of its sides for 40 seconds. Output light intensity (≥500 mW/cm²) was periodically checked using an LED radiometer (Kerr Dental Specialties, West Collins Orange, CA, USA). The specimens were left to stand for 15 minutes and then were immersed in artificial saliva¹⁷  for one month at 37°C.

Following one month, each substrate surface was wet-ground flat with 320-grit silicon carbide grinding paper corresponding to the roughness obtained by diamond bur grinding.^9,18  Each specimen was then washed with tap water for 30 seconds and blotted dry. Then the surface was etched using the 37% phosphoric acid Scotchbond (3M ESPE, St. Paul, MN, USA) for 15 seconds. The etched surface was rinsed with an oil-free air water syringe for 15 seconds and dried with air for 5 seconds from a distance of 1 cm.

Prepared substrate specimens were divided into two main groups (n=24) according to the silane primer application. Silane primer coupling agent (Clearfil Ceramic Primer Kuraray Medical Co, Osaka, Japan) was applied to one-half of the prepared substrate surfaces for 60 seconds and gently air-dried. The other half was not treated with silane. Then each group was further subdivided into two subgroups (n=12) according to the repair material, using either Filtek P90 (shade B2) with its corresponding adhesive system (P90 System Adhesive [SA]) or Filtek Z250 resin composite (shade B2) with Adper Scotchbond Multipurpose adhesive system (SBMP). Each substrate specimen was reinserted in the repair mold while the treated surface was directed upward. Adhesive systems were applied according to the manufacturers' instructions as presented in Table 1. The repairing resin composite was packed against the treated side of the Filtek P90 substrate specimen incrementally (1.5-mm thick followed by 2-mm thick). Each increment was cured for 40 seconds. A different shade was chosen for the repairing composite to enable visual identification and orientation of the repair interface during μTBS testing and failure mode analysis. Finally, each subgroup was subdivided into two categories (n=6) according to the storage periods of the repaired assembly for short-term (24 hours) and long-term (two years) in artificial saliva at 37°C. Twelve additional unrepaired substrate specimens of 4 mm diameter and 7.5 mm height were prepared from each resin composite (Filtek P90 and Filtek Z250) to test their cohesive strength^9,12  after 24 hours and two years of storage.

After storage, specimens were fixed to the cutting machine (Isomet, low-speed saw, Lake Bluff, IL, USA) and serially sectioned to obtain multiple beam-shaped sticks. From each specimen, four sticks were tested, resulting in testing 24 sticks for each experimental category. The cross-sectional area (0.6 mm² ± 0.01 mm) was confirmed with a digital caliber (Mitutoyo digital caliber, Mitutoyo Corp, Kawasaki, Japan). For μTBS testing, each stick was fixed to the testing jig attached to the universal testing machine (Lloyd LRX, Lloyd Instruments Ltd, Fareham Hants, UK) using cyanoacrylate adhesive (Rocket, Dental Ventures of America Inc, Corona, CA, USA). The sticks were stressed in tension at a cross-head speed of 0.5 mm/min until failure. The load at failure was recorded in N, and the bond strength was calculated as MPa by dividing the load by the cross-sectional area at the bonded interface. The mean and standard deviation (SD) of each group were calculated. Comparison between groups was performed using the general linear model (GLM) with μTBS as the dependent variable and the surface priming, intermediate adhesive agents/repair composites and storage periods as independent variables while taking into consideration that every specimen had generated four stick values. A Bonferroni post hoc multiple-comparison test was used when indicated. The GLM was also used to test the interaction between each of the two independent variables as well as the interaction among the three variables. A t-test was used to compare cohesive strength values of Filtek P90 and Filtek Z250 at each storage period as well as to compare the mean values of each repair category and cohesive strength values of Filtek P90 and Z250. p<0.05 was considered statistically significant. Data were analyzed using SPSS for Windows (Statistical Package for Social Sciences, release 15 for MS Windows, 2006, SPSS Inc, Chicago, IL, USA).

Failed parts of the tested sticks were mounted on an aluminum stub, sputter coated with gold, and observed under a scanning electron microscope (SEM; 515; Philips, Einhoven, the Netherlands). Failure modes were evaluated at 100× and classified into cohesive in intermediate adhesive layer or resin composite, adhesive at the interface of the substrate side, or mixed (adhesive at the interface of the substrate side and cohesive in the adhesive layer and/or in resin composite).

Two bonded sticks from each prepared specimen at each storage period were used for nanoleakage determination. Sticks were immersed in 50% ammoniacal silver nitrate solution, which was prepared according to the protocol described by Tay and others.¹⁹  The sticks were placed in the silver nitrate solution in darkness for 24 hours, rinsed thoroughly in distilled water, and immersed in photo-developing solution for eight hours under fluorescent light to reduce silver ions into metallic silver grains within voids along the bonded interface. Sticks were then polished using SiC paper of increasing grit size (1000, 1200, 2500, and 4000), rinsed with water for 30 seconds, and left to air-dry in a desiccator. Sticks were mounted on aluminum stubs, sputter coated with gold. The substrate composite-repair composite interface was analyzed using SEM (515; Philips) operated in the backscattered electron mode.

The descriptive statistics, means, and SDs for the μTBS (MPa) of the eight tested repair categories are presented in Table 3. Results of the cohesive strength values of Filtek P90 and Filtek Z250 are also shown in Table 3. GLM analysis revealed no significant effect for the silane priming application, intermediate adhesive agent/repair composite, and storage period as well as for their interactions on the μTBS values of the repaired specimens (Table 4). A t-test indicated no significant difference between the cohesive strength values of Filtek P90 and Filtek Z250 at each storage period, as presented in Table 3. All tested repair categories were statistically significantly lower than the cohesive strength of Filtek P90 and Z250 (Table 3).

Failure mode analysis is shown in Figure 2. Cohesive fracture of the substrate and the repair materials were not seen. Overall, most failures were mainly adhesive at the substrate side followed by mixed failures. Few sticks showed cohesive failure in the intermediate layer. Representative SEM images of the fracture surfaces are shown in Figures 3 and 4.

Representative images of the tested groups can be seen in Figures 5 and 6. No silver nitrate deposits were seen for the SA/Filtek P90 repair categories either with or without silane application after 24-hour storage. However, few silver nitrate deposits were seen in the SBMP/Filtek Z250 repair groups either with or without silane application after 24-hour storage. After two years of storage in artificial saliva, silver nitrate deposition was detected in all of the tested groups.

Because the dentist often has no information about the chemical composition of the existing composite, the repair of silorane-based restorations is critical. Some researchers^5,6  found that the use of silane primer might be reasonable when the composite to be repaired cannot be identified, especially when the repair system is dimethacrylate based. However, the results of the present study revealed that the use of silane primer did not improve the short- or long-term repair μTBS when a silorane-based or methacrylate-based repair system was used. Composite repair with silane application is a point of debate in many studies.^20–22  Silanes are molecules with two main functional groups: the silanol, which bonds to the silica of the composite filler, and the organofunctional group, which copolymerizes to the methacrylate of the bonding agent.^15,23  The silane also enhances the wetting of the surface for the bonding agent, which is expected to infiltrate more easily through the irregularities created by the surface roughening. Previous researchers,^22,24–27  who used silane primer in repairing methacrylate-based resin composite, proved that it had no significant effect in the repair bond strength. On the contrary, some researchers^5,10  confirmed the important role of silane primer when silorane-based resin composite was repaired with methacrylate-based adhesive/repair composite but not with the silorane-based intermediate adhesive agent/repair composite. However, it should be emphasized that these studies used a shear mode of testing as well as different methacrylate-based intermediate adhesive agents. The risk of silane priming as a step in the repair protocol is tooth substrate contamination.⁵  This is not yet clear and requires further research.

In this study, there was no significant difference between silorane-based composite repaired with a different intermediate adhesive agent/composite and that repaired with the same material. Based on these findings, the first null hypothesis was accepted. Previous studies proved that it is possible to repair a silorane-based composite with the same material^5,9,10  or even in combination with dimethacrylate-based adhesive/composite⁹  if the appropriate repair technique is used. Quartz fillers in Filtek P90 are surface treated with an oxirane-functionalized silane. Therefore, there is an expected chemical affinity between the treated fillers and the P90 System Adhesive. However, in the present study, mechanical (with finishing or polishing) and chemical (after acid etching) treatments were applied to the surface of Filtek P90, which in turn removed the functional silane from the exposed fillers. This rendered them with no affinity to adhesives. Based on this, the micromechanical and/or the chemical coupling to the resin matrix is expected to be the cause of the obtained repair bond strength of Filtek P90 and its adhesive. For SBMP, there is no chemical affinity between its components and Filtek P90; thus, micromechanical retention may contribute to the repair mechanism.⁹  Wiegand and others¹⁰  reported that repair of aged silorane requires adequate mechanical surface treatment and application of the corresponding intermediate adhesive agent/composite. Researchers agreed that the bonding of new to old composite is mainly micromechanical,^3,9,28  although chemical bonding cannot be disregarded.²⁹ 

There are no available data in the literature on the repair bond strength considered adequate enough to survive in occlusal function.³⁰  Therefore, many studies^20,30,31  have suggested that the cohesive strength of the intact unrepaired material should be taken into consideration as a control in the evaluation of the repair bond strength.²⁴  The results of this study clearly indicated that none of the repaired specimens was as strong as the mean cohesive strength of the control specimens. Nevertheless, the repair μTBS ranged from 66% to 77% of the cohesive strength of the intact unrepaired silorane-based specimens. This corresponds to previous findings, although the test materials and methodologies were different.^20,31,32 

Silorane repair μTBS of all stored repair categories were comparable and had no significant statistical difference. This result demonstrated that hydrophilicity of the intermediate adhesive agent had no effect on the silorane-based resin composite repair bond strength durability. This leads to the acceptance of the second hypothesis. There is no previously published work that investigates the long-term repair bond strength of silorane-based resin composite. Costa and others³³  found no significant difference when SBMP was used as the intermediate-adhesive agent to repair methacrylate-based resin composite. The repair bond strength did not change after 6 months of water storage.

Materials with equal bond strengths do not always fail in the same manner.³⁴  For all of the 24-hour stored repair categories, adhesive failure at the substrate side was the predominant mode of failure. After two-year storage, this percentage slightly decreased, while cohesive in the adhesive layer and mixed types of failure tended to increase. These results denote that the repair interface is still the weakest part, especially at the substrate side.⁹  After two-year immersion in artificial saliva, different parts of the repair interface may become fatigued and degraded by time. One of the interesting findings that can be interpreted from the SEM pictures is that the surface topography of the repair side of the adhesively failed sticks was almost a positive duplicate to the negative depressions present on the roughened surface of the substrate side of the failed stick. This may emphasize the previously mentioned postulation in the Mobarak⁹  study, in which resin-filled depressions go in the same line as the applied tensile forces. Hence, these depressions in the prepared surface may add to the surface area for adhesion rather than creating macromechanical retention.

The results of the present study suggest the partial acceptance of the third null hypothesis, whereby silane application did not influence the nanoleakage results. After 24 hours of storage, no silver nitrate deposition was detected when silorane-based resin composite was repaired with the same material. This finding could be explained by the relative hydrophobicity and the perfect coupling between the repaired biomaterial assembly. However, for the methacrylate-based (SBMP/Filtek Z250) repair system, silver nitrate deposition was detected at 24 hours. Also, silver nitrate deposition increased relatively after two-year storage in all groups. Thus, nanoleakage evaluation supports the SEM observation of the mode of failure percentage. It seems that differences in the material polarity play a preponderant role in silver nitrate deposition.³³  No published study tested the nanoleakage of repaired silorane-based resin composite. However, Costa and others³³  reported that adhesive interfaces of repaired methacrylate-based resin composite absorb water after long-term water storage (six months). The amount of water sorption was reported to be positively correlated with the hydrophilicity of the adhesive system.^35,36  Absorbed water molecules occupy the free volume between polymer chains and crosslinks, causing swelling of the polymer structure and leading to plasticization and softening of the resin structure.³⁷  After this relaxation process, unreacted monomers trapped in the polymer network are released to their surroundings, resulting in a higher solubility.³⁶  This might create microvoids that were likely to be filled with silver nitrate. Although these findings did not result in any reduction in composite repair bond strength, it may represent signs of early degradation. Previous work³⁸  has shown that saliva enzymes could accelerate the adhesive degradation. Further research is required to elucidate the effect of these enzymes on repair bond durability.

The results of the study suggested that a mean bond strength value could not be taken as a sole indicator of the bond quality and that an interfacial morphological evaluation may provide additional information about interfacial bonding quality.⁹  However, it has to be taken into account that nanoleakage is not necessarily associated with an impaired clinical performance. Thus, despite the importance of laboratory studies attempting to predict the performance of biomaterials, clinical trials remain the ultimate way to collect scientific evidence on the clinical effectiveness of a restorative treatment.

Within the limitations of this in vitro study, the following conclusions were drawn:

1. 1.

   Silane application as a separate step in the repair process of the silorane-based resin composite has no effect on the short- or long-term bond durability and nanoleakage.

2. 2.

   Repair bond strength of the silorane-based resin composite appeared successfully durable irrespective of the chemistry of the intermediate repair adhesive agent/composite material.

3. 3.

   Usage of different adhesive/repair materials in repairing silorane-based composite tends to show early nanoleakage.

The authors would like to thank Mr Mohamed El Shahat for the mold fabrication as well as Mrs Yomna El Wakil for her generous technical assistance and for her help in proofreading the article.

The authors of this article certify that they have no proprietary, financial, or other personal interest of any nature or kind in any product, service, and/or company that is presented in this article.