Volumetric Cuspal Deflection of Premolars Restored With Different Paste-like Bulk-fill Resin Composites Evaluated by Microcomputed Tomography

Bulk-fill resin composites have gained popularity among clinicians for posterior restorations due to improvements of materials and enhanced curing,¹  which controls the effect of curing stresses.^2,3  Although the conventional 2-mm incremental layering procedure is needed for conventional composites, advanced bulk-fill composites can be cured as a single bulk layer with a thickness of 4 to 6 mm.⁴  According to their viscosity, these materials can be classified into “flowable/base” or “paste-like/full-body” types. Due to their low amount of filler, flowable bulk-fill composites have less wear resistance; therefore, completing the top layer with a composite, which has a high amount of filler, is necessary.^5,6  The paste-like bulk-fill materials contain higher amounts of filler. These composites are particularly viscous, sculptable, and highly wear resistant, and capping of the top layer is not necessary.⁶ 

Bulk-fill composites also contain monomers, similar to conventional composites. During the polymerization reaction, the monomers approach each other,⁷  resulting in a volumetric shrinkage of composite material.⁸  Shrinkage can create stress in the composite structure, and the dental tissue can also cause marginal gaps, microleakage, and cuspal deflection.^9,10  Cuspal deflection may lead over time to microcrack propagation, enamel cracks, crazing, reduction in fracture resistance of the restored tooth, postoperative sensitivity, pulpal problems, and, in extreme cases, cusp fracture.^11,12 

In the literature, several techniques have been reported to calculate cuspal deflection in wide class II cavities with composite restorations, and values of up to 50 μm of mean cusp deformation have been reported.^13-17  These studies used microscopy, strain gauges, flexible ribbons, linear variable differential transformers, and trans-tooth illumination techniques. With these methodologies, cuspal deflection can be measured in one dimension, giving a linear result. Two- and three-dimensional (2D and 3D) microcomputed tomography (micro-CT) allows evaluation of the restorative material itself and adaptation to the dental structures with high spatial resolution.¹⁸  This technique has proved its adequacy to evaluate the polymerization contraction vectors,¹⁹  and it can represent both the pattern and the volume of polymerization shrinkage²⁰  and calculate the shrinkage volume of the resin composites.^18,21  Although a few studies have examined cuspal deflection using micro-CT, no previous study has analyzed the 3D volumetric cuspal deflection of premolars restored with different paste-like bulk-fill composites using micro-CT.

The aim of this study was to measure the volumetric cuspal deflection of premolars restored with different paste-like bulk-fill resin composites using micro-CT. The null hypothesis was that paste-like bulk-fill resin composites would not generate different amounts of cuspal deflection from conventional resin composites.

A total of 35 freshly extracted human maxillary second premolars that were free from caries, hypoplastic defects, or cracks were selected for this study. The maximum buccal-palatal width (BPW) and the maximum mesio-distal width (MDW) of each premolar were measured prior to any work using a digital caliper (Mitutoyo, Kawasaki, Japan) with a tolerance of 10 μm, and the teeth were distributed into five groups (n=7) such that the variance of the mean BPW and MDW between groups was less than 5%. Teeth were embedded in a cube-shaped plastic mold 2 cm in diameter using chemically cured acrylic resin, and the mold extended 2 mm cervical to the cemento-enamel junction to simulate the position of the tooth in the alveolar bone and to prevent reinforcement of the crown by the base. Four spheres were placed around the acrylic base, which functioned as stable reference areas, as shown in Figure 1.

Standardized large MOD cavities were prepared in each premolar with a bucco-lingual width of 4 mm and a cavity depth of 4 mm measured from the palatal cusp tip. The dimensions of each cavity preparation were verified using a digital caliper. The cavities were prepared without proximal boxes to minimize the preparation variation.²²  The buccal and lingual walls of each cavity were prepared parallel to each other.^22,23  Cavosurface margins were prepared without beveling. Cavities were prepared with a straight diamond bur with a rounded end (#881H-012, Hager & Meisinger GmbH, Neuss, Germany) using a high-speed hand piece with copious air-water spray. The bur was changed after each cavity preparation. Teeth with pulp exposure were excluded from the study.

All samples were scanned immediately after cavity preparation using a micro-CT system (Bruker Skyscan 1275, Kontich, Belgium). The teeth were positioned with the buccal surface facing the X-ray tube to ensure standardization. The scanning conditions were as follows: 100 kVp, 120 mA, 0.5-mm Al/Cu filter, 9.1-μm pixel size, and rotation at 0.1 steps. Flat field calibration was carried out prior to each scan; moreover, the manufacturer's protocol was followed to minimize ring artifacts. Each specimen was rotated 360° during the five-minute integration time. The mean time of scanning was approximately 70 minutes. The beam hardening correction and other parameters, including optimal contrast, were set according to the manufacturer's instructions.

After the initial micro-CT scanning, to eliminate the dehydration caused by the dry environment of the scanning device, the samples were rehydrated for 30 minutes in distilled water, as in the study by Shirani and others,²⁴  and then restorative procedures were performed. Four groups received different paste-like bulk-fill composites—Beautifil-Bulk Restorative (BBR), Admira Fusion x-tra (AFX), x-tra fill (XTF), and Sonic Fill (SF)—and the control group received a conventional universal composite and Clearfil Majesty Esthetic (CME) (Table 1).

Metal matrix bands (Adapt SuperCap Matrix no 2182, Kerr-Hawe, Bioggio, Switzerland) were seated for each group. The enamel was etched for 15 seconds with 35% orthophosphoric acid (K-ETCHANT Syringe, Kuraray Noritake Dental Inc, Okayama, Japan), rinsed for 10 seconds, and then dried with cotton pellets, leaving the surface moist, according to the manufacturer's directions. This procedure was followed by the application of universal adhesive (Single Bond Universal, 3M ESPE, St Paul, MN, USA) to the entire cavity with a rubbing action for 20 seconds. The excess solvent was evaporated with a gentle air spray for five seconds. The bonding agent was cured for 10 seconds with an LED curing light (Elipar DeepCure-S LED Curing Light, 3M ESPE). The output power was periodically monitored with the radiometer of the curing unit to ensure an intensity of at least 1000 mW/cm². The four bulk-fill composites were applied to the prepared cavities in single-bulk increments, while the conventional composite was applied in three horizontal increments. Each layer was cured according to the manufacturers' instructions (Table 1).

Immediately after the restorative procedure, each tooth was scanned by micro-CT in the same manner as the initial scanning. NRecon software (ver. 1.7.4.2, SkyScan, Brüker, Kontich, Belgium) and CTAn (ver. 1.18.4.0, SkyScan, Brüker) were used for imaging and measurement of the samples. For reconstruction, the algorithm by Feldkamp and others²⁵  was used to achieve axial, 2D slices. For the reconstruction parameters, ring artifact correction and smoothing were fixed at zero, and the beam artifact correction was fixed at 50%. Cross images were then reconstructed from the whole volume using micro-CT. The reconstructed images were superimposed using DataViewer software (ver. 1.5.6.2, SkyScan, Brüker). Reference points that were distant from the area affected by shrinkage were selected to align the different images of the prepared restored teeth. The spherical balls that were attached before scanning were used for this purpose. The prepared tooth image (reference) and the restored tooth image (target) were superimposed, and the difference in volume was determined by image subtraction. The resulting image represented the volume of the cuspal deflection caused by the polymerization shrinkage of the resin composite restoration (Figures 2 and 3).

CTAn software was used for visualization and calculation of this subtracted image. A region of interest (ROI) was drawn to include the entire specimen within the sample using CTAn software, and all specifications of the program were used to analyze the microarchitecture of each sample. Once the appropriate ROI (in volume) was selected, “binary images” were obtained, and an appropriate grayscale threshold was selected to distinguish the enamel, dentin, and composite fillings. A thresholding (binarization) process was used, which entailed processing the range of gray levels to obtain a superimposed image of only black/white pixels. Despeckling was also performed to remove white speckles in the 2D image that were less than 10 voxels. The buccal and palatal regions of each restoration were evaluated separately in terms of cuspal deflection. The total cuspal deflection volume values are expressed in mm³.

The volume of cuspal deflection after restoration was analyzed using the Shapiro-Wilk test for normality and equality of variances (the Levene test), followed by a parametric statistical test. One-way analysis of variance (ANOVA) was used to compare the effect of the composite resin, and multiple comparisons were performed using the Tukey test. All tests used a significance level of α = 0.05, and all analyses were conducted with the statistical program GraphPad Prism (ver. 6.0).

The cuspal deflection values for the tested composites are shown in Table 2. One-way ANOVA showed that the resin composite type had a significant effect on cuspal deflection (p<0.0001).

Multiple comparisons (Table 2) showed that teeth restored with the conventional paste-like composite and CME (control) had significantly different cuspal deflection than those filled with paste-like bulk-fill composites (p<0.05). Among bulk-fill composites, a significant difference was observed between BBR and AFX (p<0.05).

This study investigated the effect of four types of paste-like bulk-fill composites on the volumetric cuspal deflection of maxillary premolar teeth compared to a conventional paste-like composite. Due to polymerization shrinkage, all tested composite groups showed an inward deflection of the cusps, which is consistent with the results of other studies.^14,23,26-30  All paste-like bulk-fill resin composites showed less volumetric cuspal deflection than the conventional paste-like composite group. Thus, the null hypothesis was rejected.

The amount of cuspal deflection is directly related to the amount of tooth structure loss.³¹  Therefore, in the present study, large MOD cavity preparations were performed on maxillary premolar teeth to weaken tooth structure and promote cuspal deflection. The dimensional differences of the premolar teeth used in the study were set to a maximum of 5% for standardization and to enable comparison with other cuspal deflection studies.²⁸ 

Although inexpensive and simple methods are available to measure cuspal deflection, these methods measure only linear values between two points on the tooth structure.³¹  However, micro-CT is useful to evaluate the volumetric differences.²⁹  Micro-CT enables the visualization of effects across the entire end and can analyze the boundary conditions of the exact tooth cavity.³¹  In addition, Oliveira and others²⁹  showed a strong correlation between cuspal deflection values measured using the micro-CT method and those measured using the strain gauge method. Therefore, cuspal deflection was measured by micro-CT in this study.

The difference between the cuspal deflection values of the bulk-fill composites and the conventional composite in the present study could be due to two factors or a combination thereof. One factor was the chemical composition of the different composites, and the other was the incremental (three layers and three sessions of light curing) restorative procedure of the conventional composite. The results of this study are consistent with those of other studies comparing paste-like bulk-fill composites and conventional paste-like composites.^23,29,32  In contrast, a study by Tsujimoto and others³⁰  reported that paste-like bulk-fill resin composites showed cuspal deflection similar to conventional resin composites. That study used aluminum blocks instead of extracted teeth and investigated the simulated cuspal deflection of paste-like bulk-fill and conventional resin composites with a micrometer and a confocal laser-scanning microscope. The discrepancy between our study and their results might be due to the difference in the methodology or the difference in the type of composite that was compared. Another study investigated the cuspal deflection of premolars restored with three paste-like bulk-fill composite resins and one low-shrinkage silorane-based restorative material.¹⁴  The researchers reported that silorane-based composites caused less cuspal deflection than paste-like bulk-fill composites. Since cuspal deflection is directly caused by polymerization shrinkage, that study clearly demonstrates that the limited shrinkage behavior of the silorane-based composite caused less cusp movement than methacrylate-based composites.

Another finding of the current study was that BBR showed higher volumetric cuspal deflection than AFX among the bulk-fill composites. This difference might be related to the different chemical compositions of these resins. AFX is an Ormocer-based composite, while BBR contains many different monomers. These composites also contain different types and amounts of fillers.

Although composites are termed bulk fills, these commercially available composites differ in their chemical composition. To provide a 4-mm polymerization depth, every manufacturer organizes its composite structure with different types and quantities of monomers; different types, amounts, shapes, and sizes of fillers; and different photoinitiators,^5,6  and these components differ in each composite. These varied structures also require different amounts of curing light energy for adequate polymerization.³³  Because many variables exist, it is difficult to compare these composites with each other, which limits the studies that can be performed.

This study has some limitations. The major limitation involved the insertion technique and total polymerization energy. Conventional paste-like composite cannot be used for 4-mm bulk polymerization and was applied by an incremental technique. The polymerization time varied for composite groups according to the manufacturer's instructions. Therefore, the total energy applied to the composite groups also varied. Instead of applying equal energy to all composites, each composite was cured according to the manufacturer's specified curing time and energy instructions to simulate clinical conditions. These conditions might have affected the results. The second limitation was that composite resins absorb water in the oral environment, which causes hygroscopic expansion and can neutralize the negative effects of polymerization shrinkage and relieve cuspal deflection.^29,34  Since the early effects of the resin materials were examined, no hygroscopic expansion occurred this study. The third limitation of the study was the cavity type and dimensions. Currently, the use of adhesive restorations results in smaller cavity preparations. Large MOD cavities are usually observed in the replacement of amalgam fillings or teeth undergoing endodontic treatment due to excess caries. Therefore, the type of cavity used in this study is not a common cavity preparation, but to better observe the shrinkage properties of the materials, these large cavities with weak and stretchable walls were used in the study, similar to previous studies.^27,28  The fourth limitation of this study was the microcrack evaluation. Although enamel fractures are one of the most important results of polymerization shrinkage, their detection methods include subjective observations, such as transillumination.

Many studies have examined the cuspal deflection of bulk-fill composites in vitro. In vitro studies provide information about the general characteristics of the materials, but their interpretations of the intraoral environment are based on speculation. Studies are needed that evaluate cuspal deflection with an intraoral scanner or similar devices in an intraoral environment and that examine the early and long-term effects by evaluating both patient complaints and restoration status using intraoral evaluation criteria.

Within the limitations of this study, it can be concluded that paste-like bulk-fill resin composites resulted in significantly less cuspal deflection than the conventional paste-like resin composite tested. Among the bulk-fill resin composites, BBR produced greater volumetric cuspal deflection than AFX.

This study was conducted in accordance with all the provisions of the local human subjects oversight committee guidelines and policies of the Ankara University Faculty of Dentistry Ethics Committee. The approval code for this study is 2019/01 No:01/05.

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.