Highlights
- •
Shrinkage of bulk-fill and low-shrinkage composites was comparable to conventional composites.
- •
Strain distribution across depth differed in 4-mm and 2-mm layered composites.
- •
Marginal integrity was more often compromised along the gingival than axial dentin wall.
Abstract
Objective
To compare strain and displacement of sculptable bulk-fill, low-shrinkage and conventional composites as well as dye penetration along the dentin-restoration interface.
Methods
Modified Class II cavities (N = 5/group) were filled with sculptable bulk-fill (Filtek Bulk Fill Posterior, 3M ESPE; Tetric EvoCeram Bulk Fill, Ivoclar Vivadent; fiber-reinforced EverX Posterior, GC; giomer Beautifil Bulk, Schofu), low-shrinkage (Kalore, GC), nanohybrid (Tetric EvoCeram, Ivoclar Vivadent) or microhybrid (Filtek Z250, 3M ESPE) composites. Strain and displacement were determined using the 3D digital image correlation method based on two cameras with 1 μm displacement sensitivity and 1600 × 1200 pixel resolution (Aramis, GOM). Microleakage along dentin axial and gingival cavity walls was measured under a stereomicroscope using a different set of teeth (N = 8/group). Data were analyzed using analyses of variance with Tukey’s post-test, Pearson correlation and paired t -test ( α = 0.05).
Results
Strain of TEC Bulk, Filtek Bulk, Beautifil Bulk and Kalore was in the range of 1–1.5%. EverX and control composites showed 1.5–2% strain. Axial displacements were between 5 μm and 30 μm. The least strain was identified at 2 mm below the occlusal surface in 4-mm but not in 2-mm layered composites. Greater microleakage occurred along the gingival than axial wall (p < 0.05). No correlation was found between strain/displacements and microleakage axially (r 2 = 0.082, p = 0.821; r 2 = −0.2, p = 0.605, respectively) or gingivally (r 2 = −0.126, p = 0.729, r 2 = −0.278, p = 0.469, respectively).
Significance
Strain i.e. volumetric shrinkage of sculptable bulk-fill and low-shrinkage composites was comparable to control composites but strain distribution across restoration depth differed. Marginal integrity was more compromised along the gingival than axial dentin wall.
1
Introduction
Bulk-fill composites are intended for posterior teeth restoration in 4–5 mm thick layers. In general, bulk-fill composites are a heterogeneous group consisting of (1) flowable or low-viscosity materials, mainly indicated for restoring dentin but requiring a top layer of conventional composite, and (2) sculptable, paste-like or high-viscosity materials, which may be used to restore the entire cavity. An exception is a high-viscosity fiber-reinforced bulk-fill composite EverX Posterior (GC) which requires a capping layer due to unfavorable esthetics despite favorable mechanical properties . The so-called ‘giomers’, marketed as nanohybrid, fluoride release and recharge composites, are also in the sculptable bulk-fill group. Surface pre-reacted glass (S-PRG) fillers in giomers are produced in reaction of fluoride-containing glass with water solutions of polyacids.
A clear advantage of bulk-fill composites is reduced clinical working time, especially for large cavities requiring a number of 2 mm thick layers of conventional composites. Among main concerns with this ‘class’ of composites are polymerization shrinkage and marginal integrity failure. Polymerization shrinkage of bulk-fill composites was reported to be in the range of 2–3% with the accompanying shrinkage stress between 1.7 and 2.4 MPa, which was sufficient to cause interfacial debonding of bulk-fill restorations . Lower shrinkage was reported for high-viscosity bulk-fill composites when adhesively bonded to enamel and dentin than without adhesives .
Low-shrinkage composites generally contain long-chain cross-linking monomers intended to compensate for a reduction of inter-monomer distance within a polymer compared to shorter cross-linking monomers. Commercially available low-shrinkage composites containing urethane-based or silorane cross-linking monomers have shown similar or lower shrinkage than selected conventional dimethacrylate-based composites . An experimental low-shrinkage composite based on a urethane-based monomer roughly twice the molecular weight of bisphenol A glycidyl methacrylate (BisGMA) has shown 2–3 times lower shrinkage than BisGMA-based composite . Besides monomer chemistry, filler chemistry and content are also variable in low-shrinkage composites, making it impossible to represent low-shrinkage composites by any one material, similarly as with any other ‘class’ of composite materials. Selecting one or a few composites from a certain ‘class’ does not provide conclusive evidence of the behavior of the whole ‘class’, yet it is a generally accepted approach in testing dental materials.
In previous studies, shrinkage of bulk-fill composites was measured linearly using a strain gage method , a modified bonded disc method , a custom-made linometer or a Kaman linometer . Rarely is shrinkage measured using three-dimensional (3D) methods such as micro-computed tomography (μCT) providing data on internal deformation fields of the shrinking polymer. Though highly informative and precise, as well as non-destructive, μCT measurements are often time-consuming . Another way to measure 3D polymerization shrinkage of dental composites is the digital image correlation method based on two cameras . In this non-destructive and high-resolution method, composite specimens are photographed before and after light-curing. Images, acquired far quicker than with μCT, are imported in specialist software to calculate von Mises strain and displacement. In the case of dental composites von Mises strain equals volumetric shrinkage. Unlike linear methods, digital image correlation offers a detailed analysis of local deformation fields and allows zones of greater and lower shrinkage and displacements to be differentiated.
Though a number of studies tested marginal integrity, only a few high-viscosity and one low-viscosity bulk-fills were included in previous studies . Better marginal integrity of bulk-fill composites was reported at enamel than dentin and following a ‘total-etch’ than a ‘self-etch’ adhesive approach . Similar gap formation was reported for bulk-fill composites placed either as bulk or layered . In previous studies, marginal integrity of bulk-fill composites was tested using dye penetration in Class I , Class II or Class V restorations and replicas of Class II restorations for scanning electron microscopy .
The aim of the study was to measure (1) von Mises strain i.e. volumetric shrinkage and displacement of sculptable bulk-fill, low-shrinkage and conventional composites upon polymerization and (2) axial and gingival microleakage along the restoration-dentin interface. The null hypotheses were that there are no differences in strain, displacements and microleakage among the tested composite.
2
Materials and methods
2.1
Specimen preparation
Intact, human third molars extracted for orthodontic reasons were cleaned of debris and stored in 0.02% thymol at +4 °C. Ethical approval to collect such teeth for research purposes was granted by the Ethics Committee of the University of Pristina at Kosovska Mitrovica. Each tooth was embedded in super-hard gypsum up to the enamel–cementum junction. Occlusal one third was sectioned parallel to the occlusal plane using a slow-speed diamond saw (Isomet 4000; Buehler, Lake Bluff, IL, USA) to remove the cusps and expose flat dentin. Class II (‘slot’) cavity was prepared using a round-end cylindrical bur in a high-speed handpiece. Cavity dimensions, verified with a digital caliper (d = 0.01 mm), were 4 mm occlusal–gingival height, 4 mm vestibular–oral width and 2 mm depth toward the pulp.
Composites used in the study with their respective adhesive systems are presented in Table 1 . Each adhesive was used according to manufacturer’s instructions. Adhese Universal was applied following a ‘total-etch’ or a ‘self-etch’ protocol to bond TEC Bulk and TEC. These groups are indicated as ‘TEC Bulk TE’ and ‘TEC TE’ or ‘TEC Bulk SE’ and ‘TEC SE’, respectively. Single Bond Universal was applied following a ‘total-etch’ or a ‘self-etch’ protocol to bond Filtek Bulk. These groups are named ‘Filtek Bulk TE’ and ‘Filtek Bulk SE’, respectively. ‘Z250’ group was bonded with Single Bond Universal following only the ‘total-etch’ protocol. FL Bond II was applied following a ‘self-etch’ protocol to bond ‘Beautifil Bulk’ group. G-aenial bond was also applied following a ‘self-etch’ protocol to bond ‘Kalore’ and ‘EverX’ groups. Adhesives were cured for 20 s using an LED light-curing unit (LEdition, Ivoclar Vivadent, Schaan, Liechtenstein) with an output intensity of 500–600 mW/cm 2 . Bulk-fill composites (TEC Bulk, Filtek Bulk, Beautifil Bulk and EverX) were applied in 4-mm thick single layers and light-cured for 40 s. Low-shrinkage Kalore and control composites TEC and Z250 were applied in two 2-mm thick layers, each separately cured for 40 s using the same light-curing unit.
| Materials | Manufacturer | Type | Composition | Layer thickness (composite) | |
|---|---|---|---|---|---|
| Application protocol (adhesive) | |||||
| Composite | Tetric EvoCeram Bulk Fill (Code: TEC Bulk) | Ivoclar Vivadent, Schaan, Liechtenstein | Nanohybrid bulk-fill | BisGMA, UDMA, BisEMA, Fillers, barium aluminium silicate glass fillers, ytterbium fluoride and spherical mixed oxide, Ivocerin ® , prepolymer, additives, initiators, stabilizers, pigments | 4 mm |
| Filtek™ Bulk Fill Posterior Restorative (Filtek Bulk) | 3M ESPE, St. Paul, MN, USA | Nanofilled bulk-fill | Aromatic UDMA, UDMA, silica, DDDMA, silane treated ceramic, pentanedioic acid, 2,2-dimethyl-4-methylene-reaction products with glycidyl methacrylate, EDMAB, benzotriazol, titanium dioxide | 4–5 mm | |
| Beautifil Bulk Restorative (Beautifil Bulk) | SHOFU, Tokyo, Japan | Giomer bulk-fill | Bis-GMA, UDMA, Bis-MPEPP, TEGDMA, S-PRG filler based on F-B-Al-silicate glass, initiator, pigments | 4 mm | |
| EverX Posterior (EverX) | GC, Tokyo, Japan | Fiber-reinforced bulk-fill | Bis-GMA, TEGDMA, PMMA, SiO 2 , barium glass, glass fibers, initiator | 4 mm | |
| Kalore (Kalore) | GC, Tokyo, Japan | Nanohybrid low-shrinkage | UDMA, BisEMA, BHT, dimethacrylate, DX-511 co-monomers, fluoroaluminosilicate glass, pre-polymerized filler, strontium glass, SiO 2 , initiator, pigments | 2 mm | |
| Tetric EvoCeram (TEC) | Ivoclar Vivadent, Schaan, Liechtenstein | Nanohybrid composite (control) | BisGMA, UDMA, BisEMA, barium glass filler, ytterbium trifluoride, mixed oxide, prepolymers, additives, catalysts, stabilizers and pigments | 2 mm | |
| Filtek Z250 (Z250) | 3M ESPE, St. Paul, MN, USA | Microhybrid composite (control) | BisGMA, TEGDMA, UDMA, BisEMA6, aluminum oxide, EDMAB, silane treated ceramic, initiators, stabilizers, pigments | 2 mm | |
| Adhesive | Adhese Universal | Ivoclar Vivadent, Schaan, Liechtenstein | Single-component universal | HEMA, BisGMA, ethanol, 1,10-decandiol dimethacrylate, methacrylated phosphoric acid ester, CQ, DMAEMA | Total-etch / Self-etch |
| Single Bond Universal | 3M ESPE, St. Paul, MN, USA | Single-component universal | MDP, dimethacrylate resins, HEMA, Vitrebond™ copolymer, filler, ethanol, water, initiators, silane | Total-etch / Self-etch | |
| FL Bond II | SHOFU, Tokyo, Japan | Two-step self-etch | HEMA, UDMA, TEGDMA, glass powder | Self-etch | |
| G-aenial Bond | GC, Tokyo, Japan | One-step self-etch | Hydroxy-dimethacryloxypropane, methacryloyloxydecyl dihydrogen phosphate, ethylenedioxydiethyl dimethacrylate, diphenyl-trimethylbenzoylphosphine oxide, acetone | Self-etch | |
2.2
Volumetric shrinkage and displacement measurements
Digital image correlation based on a previously described two-camera system (Aramis, GOM GmbH, Braunschweig, Germany) was used to determine volumetric shrinkage and displacements of the tested materials. The cameras had 1600 × 1200 pixel resolution and 12 Hz maximum frame rate. The cameras were first calibrated using the calibration panel provided by the manufacturer. The proximal surface of each composite specimen (N = 5 per group) was sprayed with a fine layer of black and white acrylic paint (Kenda Color Acrilico, Kenda Farben, Ferrera Erbognone, Italy) to produce irregular speckles for tracking by the cameras whilst the occlusal surface was protected with a Mylar strip. The paint dried within 30 s during which time the samples were kept in dark to avoid ambient light-induced polymerization. Light-curing was performed from the occlusal direction at a fixed distance of 1 mm using the same LEdition light-curing unit. The images were taken immediately before and after light-curing and further computational analysis was performed in Aramis specialist software. The experimental setup is schematically presented in Fig. 1 .
2.3
Dye penetration test
Marginal integrity was evaluated based on microleakage using a dye penetration test. Upon curing, the specimens (N = 8 per group) were covered with two layers of nail varnish except 1 mm around cavity margins. The specimens were immersed in 50 wt% ammoniacal AgNO 3 solution for 2 h, rinsed under tap water and then immersed in film developer for 6 h. Following rinsing and blot-drying with a paper towel, the specimens were sectioned using the same slow-speed diamond saw (Isomet 4000) parallel to the long tooth axis to produce 1 mm thick slices. Microleakage was determined by digitizing each slice using a CCD camera under a stereomicroscope at ×30 magnification and measuring dye penetration with a built-in millimeter scale along the axial and gingival cavity walls, both located in dentin.
2.4
Statistical analysis
Two-way ANOVA was used to test the effects of “composite” and “adhesive application” on strain and displacement of three tested composites: TEC Bulk, Filtek Bulk and TEC because the same adhesive was used following both ‘total-etch’ and ‘self-etch’ protocols. In order to fit all data in one model, the tested composites with their respected adhesives were considered ‘restorative systems’. Von Mises strain and displacement datasets for all restorative systems were analyzed using one-way analysis of variance (ANOVA) with Tukey’s post-hoc test.
Microleakage data for all tested restorative systems were analyzed using a two-way ANOVA for the factors “restorative system” and “leakage domain (axial/gingival)”. The interaction between factors was also tested. Since intra-group data related to axial and gingival microleakage were obtained from dependent specimens, intra-group comparison i.e. axial vs. gingival leakage for each restorative system was performed using a paired t -test with the Bonferroni correction to control Type I error rate at 0.05.
Pearson correlation was used to evaluate the relationship between axial and gingival microleakage as well as between strain/displacement and dye penetration. The level of significance for all analyses was set at 0.05.
Power and sample size calculations showed that 5 samples per group give 80% power in detecting differences in strain and displacement of 0.8% and 10 μm, respectively whereas 8 samples per group give 80% power in detecting 0.5 mm difference in microleakage based on the mean standard deviation values obtained in the study. As there is no conclusive evidence confirming clinically relevant values of strain, displacement and microleakage, the abovementioned values were considered acceptable.
2
Materials and methods
2.1
Specimen preparation
Intact, human third molars extracted for orthodontic reasons were cleaned of debris and stored in 0.02% thymol at +4 °C. Ethical approval to collect such teeth for research purposes was granted by the Ethics Committee of the University of Pristina at Kosovska Mitrovica. Each tooth was embedded in super-hard gypsum up to the enamel–cementum junction. Occlusal one third was sectioned parallel to the occlusal plane using a slow-speed diamond saw (Isomet 4000; Buehler, Lake Bluff, IL, USA) to remove the cusps and expose flat dentin. Class II (‘slot’) cavity was prepared using a round-end cylindrical bur in a high-speed handpiece. Cavity dimensions, verified with a digital caliper (d = 0.01 mm), were 4 mm occlusal–gingival height, 4 mm vestibular–oral width and 2 mm depth toward the pulp.
Composites used in the study with their respective adhesive systems are presented in Table 1 . Each adhesive was used according to manufacturer’s instructions. Adhese Universal was applied following a ‘total-etch’ or a ‘self-etch’ protocol to bond TEC Bulk and TEC. These groups are indicated as ‘TEC Bulk TE’ and ‘TEC TE’ or ‘TEC Bulk SE’ and ‘TEC SE’, respectively. Single Bond Universal was applied following a ‘total-etch’ or a ‘self-etch’ protocol to bond Filtek Bulk. These groups are named ‘Filtek Bulk TE’ and ‘Filtek Bulk SE’, respectively. ‘Z250’ group was bonded with Single Bond Universal following only the ‘total-etch’ protocol. FL Bond II was applied following a ‘self-etch’ protocol to bond ‘Beautifil Bulk’ group. G-aenial bond was also applied following a ‘self-etch’ protocol to bond ‘Kalore’ and ‘EverX’ groups. Adhesives were cured for 20 s using an LED light-curing unit (LEdition, Ivoclar Vivadent, Schaan, Liechtenstein) with an output intensity of 500–600 mW/cm 2 . Bulk-fill composites (TEC Bulk, Filtek Bulk, Beautifil Bulk and EverX) were applied in 4-mm thick single layers and light-cured for 40 s. Low-shrinkage Kalore and control composites TEC and Z250 were applied in two 2-mm thick layers, each separately cured for 40 s using the same light-curing unit.
