Evaluation of Microvoid and Microleakage Potential of Bulk-Fill Resin Composites in MOD Restorations

Abstract

Background: Polymerization shrinkage and microvoid formation has been a significant problem giving way to resin composite failure. The aim of this study was to evaluate the microvoid potential and microleakage of two bulk-fill and a microhybrid resin composite applied with different adhesive materials. Materials and Methods: MOD cavities were prepared in 60 endodontically treated maxillary premolars. The teeth were divided into six different groups (n = 10) according to adhesive systems (Gluma (Kulzer), OptiBond FL (Kerr) and resin composite materials EverX Posterior (GC) and Filtek One Bulk Fill (3M ESPE). An aging procedure with 5000 cycles of thermal cycling was applied. All teeth were individually scanned with micro-computed tomography. A Shapiro–Wilk test, two-way MANOVA, and Bonferroni analysis were used for statistical tests. Results: Self-etch groups showed more microleakage than total-etch groups. Minimum microleakage was observed in Filtek One Bulk Fill groups, whereas G-aenial Posterior groups showed maximum microleakage. Conclusions: Filtek One Bulk Fill (3M ESPE) resin composite was found to be more effective in reducing microvoid formation in MOD cavities when applied with total-etch adhesive systems. However, EverX Posterior (GC) and G-aenial Posterior (GC) showed less microvoid formation with self-etch adhesive systems.

1. Introduction

Due to increasing aesthetic expectations and advances in bonding methods, dentists often prefer resin composite materials for all teeth that can be treated with composite resin for both anterior and posterior tooth restorations. The clinical outcomes of these restorations are influenced by the application techniques of dental adhesives and composite resins [1].

Large voids within restorative materials can lead to reduced fracture resistance, microleakage, and marginal discoloration [2]. Proper material usage ensures successful adhesion between the restorative material and the tooth, preventing the occurrence of microfissures and voids.

To achieve an optimal resin composite restoration, it is crucial to consider risk factors at the restoration site. Despite a successful adhesive system, shrinkage can occur at the cusp, potentially leading to enamel fractures [3,4]. Among the most common causes of posterior restoration failure, secondary caries and bulk fractures can occur as a result of polymerization shrinkage, which is the main cause of failure [5,6]. Polymerization shrinkage remains a significant challenge, often contributing to resin composite failure [7]. Additionally, the dimensional stability of the restorative material plays a vital role in preventing microleakage at the interface between the tooth and the restorative material [8]. Ensuring proper handling and understanding of these factors is essential for successful dental restorations. Shrinkage stress can significantly impact the marginal stability of restorative materials, potentially leading to issues such as microleakage, secondary caries, and postoperative sensitivity. The C-factor is defined as the proportion of tissue attached to the unattached surfaces of the tooth cavity, and its value is thought to be directly related to the stress developed at the interfacial attachment site [9]. Moreover, in the case of direct resin composite restorations within large Class I and Class II cavities characterized by a high C-factor, shrinkage stress may result in microcracks and cuspal deflection within the enamel tissue, ultimately reducing fracture resistance [10,11].

Different approaches have been suggested to reduce polymerization shrinkage in dental restorations. These include applying composite resin materials by incremental technique and using dental LED lights with a soft start mode. However, incremental technique can be time-consuming and may lead to gap formation between the cavity wall and the restorative material, as well as disconnection of the bond between composite layers [12]. Since the incremental technique is time-consuming and can cause gap formation between the cavity wall and the restorative material and disconnection between the composite layers, especially in deep cavities, composite resins have been developed to overcome these disadvantages. Considering the drawbacks of conventional resin composite materials applied by incremental technique, bulk-fill resin composites have been introduced to the market in order to prevent the disadvantages that may occur, to save time and to decrease costs [13]. The key advantage of bulk-fill resin composites lies in their ability to be applied as a single bulk layer in cavities with a depth of up to 4 mm. Bulk-fill resin composite has some advantages, such as there is no need for it to be applied in layers for curing, it does not need a longer curing time, and it does not need a higher light intensity [13,14]. With all these properties, manufacturers have emphasized that bulk-fill resin composites have lower polymerization shrinkage than flowable and conventional composites [15]. The improved physical and mechanical properties of bulk-fill resin composites provide ease of use for filling deep cavities.

EverX Posterior is a fiber-reinforced composite that has been developed in recent years. The addition of fiber strengthens the structure of EverX Posterior and increases the durability of the material. EverX Posterior, which is used as a dentin replacement material, increases the fracture strength and prevents crack progression due to the short fibers in its content, especially in large and deep cavities [16]. There are very few studies in the literature evaluating the microleakage and microvoid potential of EverX Posterior, using the micro-CT device. The use of micro-computed tomography provides a detailed 3D reconstruction of teeth, restorations, and surrounding structures that can be sliced in any direction to obtain accurate results that provide information about internal geometric properties and structural parameters [17,18]. One of the advantages of micro-CT is its ability to be non-destructive, especially in temporal assessment [17].

The aim of this study was to evaluate the impact of self-etch and total-etch adhesives on microvoid formation and microleakage of two bulk-fill and a microhybrid resin composite in deep MOD cavities by using micro-CT.

The following hypotheses were determined for this study:

Hypothesis 1.

The choice of restorative material does not influence microvoid formation.

Hypothesis 2.

The choice of restorative material does not influence microleakage formation.

Hypothesis 3.

The choice of adhesive system (etch-and-rinse or self-etch) does not influence microvoid formation.

Hypothesis 4.

The choice of adhesive system (etch-and-rinse or self-etch) does not influence microleakage formation.

2. Materials and Methods

2.1. Preparation of Specimens

Ethical approval (NEU/2023/112-1697) was granted by the Near East University Scientific Research Ethics Committee on 30 March 2023, ensuring compliance with ethical standards and participant protection. The teeth used in the study were free of cracks, fractures, caries, and restorations and had almost similar buccopalatinal length and mesiodistal width. The criteria inclusion and exclusion in the tooth selection are shown in

Table 1. Inclusion and exclusion criteria for teeth selection.

Inclusion Criteria	Exclusion Criteria
Maxillary first premolar teeth	Presence of dental staining (tetracycline, trauma, fluorosis, and unknown etiology)
Maxillary first premolar teeth having two separate roots	Presence of direct and indirect restorations in the occlusal and aproximal region
Maxillary first premolar teeth, usually with a crown length of 8 mm and a root length of 14 mm	Tooth with extensive caries
Maxillary first premolar teeth with little or no caries, no fractures and cracks	Fractured and cracked teeth

. A total of 60 maxillary premolars were divided into six groups, each including 10 teeth. The teeth were individually embedded in C-type silicone impression material (Zetaplus, C-silicone putty, Zhermack, Italy) up to the enamel–cement margin. Afterwards, all groups underwent root canal treatment with the help of an endodontist in our faculty. After the access cavity was opened, the working length was determined by passively inserting a number 15 K-type file (Dentsply Maillefer, Ballaigues, Switzerland) until it reached the apical foramen, which was 0.5 mm shorter than the measured length. Root canals were prepared with ProTaper Universal rotary instruments (F2, Dentsply Maillefer, Baillagues, Switzerland). At each file change, the root canals were washed with 2 mL of 2.5% sodium hypochlorite (Cerkamed). The root canals were dried using paper points (F2, Dentsply Maillefer, Baillagues, Switzerland). After the completion of chemomechanical cleaning, the root canals were filled with Endoseal (Prevest DenPro, Jammu, India) root canal paste by lateral compaction technique using gutta-percha (F2, Dentsply Maillefer, Baillagues, Switzerland) points. MOD cavities 5 mm deep and 3 mm wide on the proximal and occlusal surfaces of the teeth were prepared by the same operator using a high-speed handpiece with air/water spray and diamond fissure burs. The proximal walls were straightened with diamond flame-tipped burs. Depth and width were checked with a digital compass to check the standards of the cavities.

After cavity preparation, the teeth were divided into 6 different groups according to the adhesive systems and resin composite materials to be used. The Tofflemire matrix system was used to restore anatomic proximal contours before resin composite application. The study method is shown in Figure 1. Two bulk-fill resin composites. Filtek One Bulk Fill (3M ESPE) and EverX Posterior (GC), and a conventional microhybrid resin composite, G-aenial Posterior (GC), were used in the study. Self-etch (Gluma, Kulzer) and total-etch (OptiBond FL, Kerr) adhesive systems were used for bonding. The composites used in the study and their composition are shown in

Table 2. Composition of resin composite materials used in the study.

Materials	Manufacturer	Type	Composition
EverX Posterior (EP)	GC Corporation Tokyo, Japan	Bulk-fill	Bis-GMA, PMMA, TEGDMA, Barium glass fillers, and silanated e-glass fibers.
Filtek One Bulk Fill (FBF)	3M ESPE, St. Paul, MN, USA	Bulk-fill	AUDMA, AFM, diurethane-DMA and 1,12-dodecane-DMA, ytterbium trifluoride, zironia/slica
G-aenial Posterior (GP)	GC Corporation Tokyo, Japan	Conventional Microhybrid	UDMA, dimethacrylates, pre-polymerized fillers, slica, strontium and lanthanide flouride

. All materials were used according to the manufacturer’s instructions.

Groups (n = 10) were planned as follows:

G-aenial Posterior—Self-etch group (control group): After air drying the cavity floor and walls, the Tofflemire matrix system was adapted to the tooth in such a way that there were no openings at the edges of the cavity. A one-step adhesive system (Gluma, Heraeus Kulzer, South Bend, India) was applied to the cavity surface with a disposable applicator for 20 s and air-dried until a uniform surface was formed. The bonding agent was polymerized by an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² for 10 s. The proximal wall was first formed with microhybrid composite resin material G-aenial Posterior (GC, Tokyo, Japan) with a thickness of 1 mm and polymerized from the occlusal surface with an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² for 20 s. Conventional microhybrid resin composite (G-aenial Posterior, GC, Tokyo, Japan) was then placed in the cavity with a thickness of 2 mm using the incremental technique and polymerized with an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² from the occlusal surface for 20 s for each increment. Finally, the proximal surfaces were polymerized using an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² for 20 s.

G-aenial Posterior—Total-etch group (control group): In total, 37.5% phosphoric acid was applied to all cavity surfaces for 15 s. During the removal of phosphoric acid from the cavity surface, excess acid was removed using cotton pellets. The cavity surface was then rinsed with water for 15 s. Primer (Optibond FL, Kerr, Orange, CA, USA) was applied with a disposable applicator for 15 s and air-dried until a uniform surface was obtained. Then, the adhesive (Optibond FL, Kerr, Orange, CA, USA) was applied for 15 s, allowed to dry until a uniform surface was obtained and cured with an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² for 10 s. A resin composite was applied as in group 1.

EverX Posterior—Self-etch group: The adhesive system procedure was applied as in group 1. The approximal wall was formed with SFRC (short-fiber-reinforced composite) resin (EverX Posterior, GC, Tokyo, Japan) and polymerized from the occlusal surface with an LED curing light at 1000–1200 mW/cm² for 20 s. A 3 mm thick layer of EverX Posterior was placed and polymerized from the occlusal surface with an LED curing light at 1000–1200 mW/cm² for 20 s. Microhybrid composite resin material (G-aenial Posterior, GC, Tokyo, Japan) was applied to the remaining part of the cavity and polymerized with an LED curing light at 1000–1200 mW/cm² for 20 s from the occlusal surface. Finally, light was applied from the approximal surfaces and polymerized with an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² for 20 s. The application procedure is shown in Figure 2, diagram A.

EverX Posterior—Total-etch group: The adhesive system procedure was applied as in group 2. A resin composite was applied as in group 3. The application procedure is shown in Figure 2, diagram B.

Filtek One Bulk Fill—Self-etch group: The adhesive system procedure was applied as in group 1. The proximal walls were first formed with bulk-fill composite resin material Filtek One Bulk Fill (3M ESPE, St. Paul, MN, USA) and cured from the occlusal surface with an LED curing light at 1000–1200 mW/cm² for 20 s. Bulk-fill composite resin material Filtek One Bulk Fill (3M ESPE, St. Paul, MN, USA) was applied to all remaining surfaces of the cavity by bulk technique and cured from the occlusal surface with an LED curing light (B-Cure Woodpecker, Guilin, China) at 1000–1200 mW/cm² for 20 s. Finally, a curing light was applied from the approximal surfaces at 1000–1200 mW/cm² for 20 s.

Filtek One Bulk Fill—Total-etch group: The adhesive system procedure was applied as in group 2. Resin composite was applied as in group 5.

2.2. Thermal Cycling

To simulate the samples in vivo for approximately 6 months, an aging procedure was performed with a thermal cycler (SD Mechatronik Thermocycler, SD Mechatronik, Westerham, Germany) for 5000 cycles at Ankara University Faculty of Dentistry [19,20]. Each group was placed and removed in the thermal cycler on the same day and time. Each group was placed separately in thin and transparent socks, and a color code was determined for each group to mix up the groups. The socks were tied with threads that suited the color code. Thermal cycling was performed (5–55 °C) with a dwell time of 25 s. The test specimens were transferred between the two temperatures with a transfer time of 10 s.

2.3. Specimen Preparation for Microleakage and Microvoid Evalutation

A 50% ammonia silver nitrate (AgH₃N₂O₃) solution was prepared at Ankara University, Faculty of Pharmacy, and used to determine the microleakage. After the aging procedure, all specimens were coated with 2 coats of nail polish, leaving 1 mm around the restoration, and kept in ammonia silver nitrate solution in dark conditions for 12 h. Each sample was kept under running water for 2 min. Then, all teeth were immersed in a photo development solution (Konix, Turkuaz Sağlık Hizmetleri, Istanbul, Türkiye) and exposed to daylight Kelvin lamp illumination for 8 h [21]. A toothbrush was used to remove silver debris on the tooth surfaces. Each sample was scanned using a micro-CT (Bruker Skyscan 1275, Kontich, Belgium) with a 100 kVp, 100 mA X-ray source. The samples were rotated 360° within an integration time of 5 min. The average scanning time was 1 h. Reconstructions were performed using reconstruction software (NRecon 1.6.7.2, Skyscan, Kontich, Belgium) with a modified algorithm obtained with a 3D density function based on 2D projection. The entrance of the optimal contrast limits (0–0.1) was set before the reconstruction of the teeth. Contrast limits were applied following the software instructions. The lowest limit was zero, so the intensity scale had zero origin. The maximum limit was at the top of the brightness spectrum, representing the highest intensity value. The data set in the study consisted of approximately 601 axial slice micro-CT images, each slice having 1024 × 1024 pixels and a 16-bit grey value. The CT software device (CTAn, 1.19.11.1, Skyscan, Aartselaar, Belgium) was used for 3D volumetric analysis and µCT volume of the sample. The reconstructed images were processed for further imaging (Skyscan CTVox, Skyscan, Aartselaar, Belgium, The Dataviewer, Skyscan, Aartselaar, Belgium). After reconstruction, a region of interest (ROI) including the teeth and the entire restoration was drawn using the CTAn software, which uses all the features of the program to analyze the 3D microstructure of each sample. Enamel and dentin were made more translucent, and restorations were brightened. As the software itself allows the user to ‘sculpt’ the desired volume from the 3D structure and by adjusting the brightness and opacity values, unwanted voxels can be removed before voids in and with restorations are visualized and calculated.

2.4. Evaluation of Microvoids with Micro-CT

Original grayscales were used to calculate cavities in 3D density, and a Gaussian low-pass filter was applied to reduce noise. An automated segmentation threshold was used to distinguish enamel and dentin from restorations and cavities using CT analysis software.

For the calculation of the volumes of the microvoids, an ROI was selected for each cross-section containing the entire restoration within the teeth. The microvoid and restoration volumes were measured for each specimen, and then the microvoid volume was calculated relative to the total restoration volume (mm³).

2.5. Evaluation of Microleakage with Micro-CT

CTAn (SkyScan) software was used to review and analyze the coronal images. A region of interest (ROI) covering the entire object was selected for each section to enable the calculation of AgH₃N₂O₃ penetration volume throughout the restoration. Grayscale shading thresholds were defined to differentiate dentin from restorative material and penetrated AgH₃N₂O₃. The volume of ammonia silver penetration was then realized by 3D analysis.

2.6. Statistical Analysis

The data obtained in the study were analyzed using the Statistical Package for Social Sciences 28.0 program [SPSS 28.0, community.ibm.com]. Descriptive statistical methods (mean, standard deviation, minimum, median, and maximum) were used to evaluate the data. The normal distribution of the data used in the study was tested with the Shapiro–Wilk test. The statistical significance of the effect of two variables and their interaction with more than one dependent variable was tested by two-way MANOVA analysis. The difference between the groups was tested with Bonferroni analysis.

For all analyses, the probability of Type I error was set as α = 0.05.

In the study, the assumptions of two-way MANOVA analysis were checked. Accordingly, whether the covariance matrix was homogeneous was tested with Box’s M test. According to Box’s test result, it was seen that the covariance matrices between groups were not homogeneous (Box’s M = 259.115, F = 4.171, df1 = 50, df2 = 5351.928, p = 0.000). In this case, it would be appropriate to use Hotelling’s trace statistic in multivariate test results. The homogeneity of the variances of each variable according to the groups in the variables was tested with Levene’s test. Accordingly, it was determined that the variances were homogeneous for microvoid (p = 0.051 > 0.05) but not homogeneous for AgH₃N₂O₃ (p = 0.045 < 0.05).

3. Results

According to Hotelling’s trace values in the multivariate test, the effects of composites, adhesive systems, and composites x adhesive system interaction on variables are significant (p < 0.05). Multivariate test results are shown in

Table 3. Multivariate test results.

	Multivarite Test
	Hotelling’s Trace	F	Hypothesis df	Error df	p	η2
Intercept	1605.191	20,466.184	4.000	51.000	0.000 *	0.999
Composites	20.526	128.285	8.000	100.000	0.000 *	0.911
Adhesive Systems	51.465	656.179	4.000	51.000	0.000 *	0.981
Composites x Adhesive systems	3.257	20.354	8.000	100.000	0.000 *	0.620

.

According to the test of between-subjects effects table, the variable composites have a significant effect on the variable AgH₃N₂O₃ (p = 0.000 < 0.05), while it has no significant effect on the variable void (p = 0.356 > 0.05). It was determined that there was a statistically significant difference between AgH₃N₂O₃ values according to Filtek One Bulk Fill, EverX Posterior, and G-aenial Posterior groups, and this difference originated from all groups. Accordingly, the mean value of AgH₃N₂O₃ was highest in the G-aenial Posterior group and lowest in the Filtek One Bulk Fill group. The amount of AgH₃N₂O₃ leakage in the G-aenial Posterior group is shown in Figure 3 and Figure 4. The highest AgH₃N₂O₃ leakage occurred in the G-aenial Posterior self-etch group and is shown in Figure 3. The least AgH₃N₂O₃ leakage occurred in the Filtek One Bulk Fill total-etch group and is shown in Figure 5. The variable adhesive systems had a significant effect on the variable AgH₃N₂O₃ (p = 0.000 < 0.05) but not on the variable microvoid (p = 0.467 > 0.05). It was determined that the AgH₃N₂O₃ value of the self-etch group was higher than the total-etch group. Tests of between-subjects effects results are shown in

Table 4. Tests of between-subjects effects results.

Tests of between-Subjects Effects
				F	p	η2
Corrected Model		AgH3N2O3 Volume		734.227	0.000 *	0.986
		Microvoid Volume		3.092	0.016 *	0.223
Intercept		AgH3N2O3 Volume		53,038.471	0.000 *	0.999
		Microvoid Volume		4721.965	0.000 *	0.989
Composites		Median (Min–Max)	$\overline{\mathit{x}}$ ± Sd
AgH3N2O3 Volume (mm3)	Filtek One Bulk Fill	6.5 (4.7–9.01)	6.77 ± 1.89 a	511.791	0.000 *	0.950
	EverX Posterior	7.94 (5.83–9.99)	7.97 ± 1.67 b
	G-aenial Posterior	9.27 (7.38–12.11)	9.51 ± 2.03 c
Microvoid Volume (mm3)	Filtek One Bulk Fill	52.35 (44.02–84.27)	54.47 ± 8.99	1.051	0.356	0.037
	EverX Posterior	56.67 (48–65.52)	56.25 ± 4.04
	G-aenial Posterior	54.81 (42.76–64.49)	53.45 ± 6.13
Adhesive Systems		Median (Min–Max)	$\overline{\mathit{x}}$ ± Sd
AgH3N2O3 Volume (mm3)	Self-etch	9.58 (7.84–12.11)	9.88 ± 1.26 a	2630.051	0.000 *	0.980
	Total-etch	6.4 (4.7–7.65)	6.28 ± 1.1 b
Microvoid Volume (mm3)	Self-etch	53.81 (42.76–84.27)	54.14 ± 8.08	0.537	0.467	0.010
	Total-etch	56.23 (45.23–65.52)	55.3 ± 5.01

* p < 0.05, ${\eta }^2$: partial eta square.

.

The effects of composites x adhesive systems interaction on the variables are statistically significant: AgH₃N₂O₃ (p = 0.001 < 0.05) and microvoid (p = 0.003 < 0.05). It was determined that there was a statistically significant difference between the AgH₃N₂O₃ volume values of the self-etch and total-etch groups in the Filtek One Bulk Fill, EverX Posterior, G-aenial Posterior groups, and the average value of self-etch was higher than total-etch in all groups. There was a statistically significant difference between the microvoid volume values of self-etch and total-etch groups in the Filtek One Bulk Fill and G-aenial Posterior groups. The average microvoid volume value of the Filtek One Bulk Fill self-etch group was higher than the Filtek One Bulk Fill applied with total-etch adhesive. The microvoid volume of the G-aenial Posterior self-etch group showed lower values than the G-aenial Posterior total-etch group. The G-aenial Posterior self-etch group demonstrated lower microvoid formation in comparison to the Filtek One Bulk Fill self-etch group. Microleakage and microvoid values between the groups are shown in

Table 5. Composites x adhesive systems interaction results.

Composites x Adhesive Systems
	Composites	Adhesive Systems	Median (Min–Max.)	$\overline{\mathit{x}}$ ± Sd	p	F	p	η2
AgH3N2O3 Volume (mm3)	Filtek One Bulk Fill	Self-etch	8.67 (7.84–9.01)	8.59 ± 0.35	<0.001 *	8.751	0.001 *	0.245
		Total-etch	4.97 (4.7–5.15)	4.94 ± 0.16
	EverX Posterior	Self-etch	9.58 (9.1–9.99)	9.57 ± 0.26	<0.001 *
		Total-etch	6.4 (5.83–6.78)	6.36 ± 0.28
	G-aenial Posterior	Self-etch	11.41 (10.89–12.11)	11.47 ± 0.37	<0.001 *
		Total-etch	7.55 (7.38–7.65)	7.55 ± 0.09
Microvoid Volume (mm3)	Filtek One Bulk Fill	Self-etch	55.92 (44.02–84.27)	57.45 ± 11.3	0.035 *	6.411	0.003 *	0.192
		Total-etch	50.68 (45.23–61.92)	51.49 ± 4.84
	EverX Posterior	Self-etch	55.53 (48–62.86)	55.52 ± 4.11	0.601
		Total-etch	57.17 (49.49–65.52)	56.97 ± 4.06
	G-aenial Posterior	Self-etch	48.95 (42.76–56.99)	49.45 ± 5.22	0.005 *
		Total-etch	58.23 (49.64–64.49)	57.45 ± 4.08

.

4. Discussion

Resin composites have become materials of choice for directly restoring both anterior and posterior teeth in dentistry. Dentists and patients alike seek materials that can replace missing tooth tissue while closely resembling naturel teeth. Additionally, resin composites address the issue of low fracture resistance often encountered in posterior teeth [22,23]. Modifying the particle size and morphology of composite resins to increase fracture resistance and reduce patient chair time have resulted in improved mechanical properties [22]. However, long-term clinical studies have pointed out an increase in the failure rate of resin composite restorations from 1.50% to 2.20% between 2006 and 2016. The causes of failure have been attributed to high rates of secondary caries, tooth fractures, and endodontic treatment [24].

The fracture strength, a crucial physical property of restorative materials, is intricately linked to their chemical composition. The combination of these components plays a pivotal role, as a material endowed with high fracture strength not only impedes the initiation but also hinders the progression of fractures. Consequently, the fracture strength emerges as a key determinant in ensuring the longevity of dental materials [25,26].

As there are studies that argue that the microstructure of conventional composite resins does not resemble dentin, the material used to replace dentin should have similar properties to dentin [27]. Researchers persist in their exploration of innovative materials designed to closely mimic the properties of dentin. These advancements are geared towards improving the efficacy of restorative procedures and ultimately enhancing patient satisfaction. Recent studies featured in the literature are dedicated to reinforcing the remaining tooth structure. In pursuit of this objective, these studies have incorporated fiber-reinforced composites into their investigations, aiming to fortify and support the structural integrity of the tooth [28,29]. Short-fiber-reinforced bulk-fill resin composites have been developed that are claimed to overcome the weak mechanical properties of conventional resin composites. During the restoration of the cavity, a biomimetic approach may be used with a short-fiber-reinforced bulk-fill resin composite as a dentin replacement and a conventional resin composite material as the enamel layer [27]. In parallel with Keulemans, Garoushi, and Lassila (2017), a short-fiber-reinforced composite as a dentin replacement was used in this study. EverX Posterior contains short-fiber particles, and the fiber particles may be exposed during polishing of the restoration. Hence, the final layer was completed using a conventional composite resin, following the manufacturer’s instructions, in accordance with a similar research study [16].

The recently introduced bulk-fill composite resins are promising, but current scientific research on their mechanical and clinical potential remains limited. Therefore, more research is needed, especially regarding their internal structure and polymerization shrinkage stress [30].

During the application of restorative material, gaps that form between the material and the tooth, as well as within the material itself, can lead to reduced fracture strength. These gaps, influenced by the adhesive system and type of restorative material used, are also responsible for marginal discoloration and microleakage. The magnitude of volumetric shrinkage and the amount of stress generated during the polymerization reaction of composite resins are the main factors causing poor marginal adaptation, microleakage, postoperative pain, and secondary caries [31]. Tsujimoto et al. (2016) emphasized that volumetric shrinkage starts immediately after light irradiation and continues even after discontinuation. The continuation of shrinkage after the completion of polymerization may be due to the polymerization reaction of the monomers [32]. It is important to use appropriate dental adhesives and restorative materials according to the width and depth of the cavity. By using the appropriate adhesive system and restorative materials, a successful bond between the restorative material and the tooth can be achieved, and microvoid formation can be prevented [2]. In the present study, micro-CT, a widely preferred powerful 3D imaging technique for polymerization shrinkage and microleakage studies, was used [2,17]. Micro-CT acquires images based on the principle of absorption or phase contrast effect [33]. In our study, no statistical difference was found between conventional microhybrid, bulk-fill, and fiber-reinforced bulk-fill resin composites in terms of void formation. However, in the self-etch groups, bulk-fill composite resin showed more microvoids than conventional microhybrid composite. Therefore, Hypothesis 1, “The choice of restorative material does not affect microvoid formation” is partially accepted. Tekçe et al. (2021) used Estelite Posterior, Estelite Flow Quick High Flow, Estelite Flow Quick High Flow, Estelite Bulk Fill Flow, and EverX Posterior on endodontically treated teeth and analyzed microvoid formation with a micro-CT device. The result of the study pointed out that flowable resin composites had more microvoid formation than conventional composites and composites containing fiber. Tekçe et al. (2021) assumed that this result may be due to the undue application of flowable resin composites [16]. Any mixing movement and placement of the flowable composite by violent spraying causes void formation [34]. It has been shown that the propensity for void formation is significantly different between various flowable composites and this is material-dependent [35]. Tekçe et al. (2021) pointed out that the lowest void volume was found in fiber-reinforced composite EverX Posterior, followed by conventional microhybrid composite Estelite Posterior by micro-CT measurements. In our study, there was no statistical difference between the conventional microhybrid resin composite and the fiber-reinforced bulk-fill resin composite in terms of void formation.

Yu et al. (2017) showed that the shrinkage of bulk-fill resin composites was between 1.5% and 3.4%, while this range was between 2.1% and 4.3% for conventional resin composites [36]. In our study, G-aenial Posterior, a conventional resin composite material, showed more microleakage than bulk-fill composites for both self-etch and total-etch groups. The fact that the cavities were 5 mm deep and the placement of conventional resin composite in 2 mm layers in the cavity caused adaptation problems can be considered as the reasons for the microleakage.

Microleakage in resin-based restorations has always been one of the biggest problems [37]. Poggio et al. (2012) stated that self-etch adhesive systems were more successful than total-etch adhesive systems. As a result of the study, the highest microleakage was observed in the One-up Bond F Plus group, followed by Palfique bond [38]. They stated that the reason why One-up Bond F Plus showed the highest leakage was that it does not contain HEMA (hydroxyethyl methacrylate) [39]. Since the hydrophilicity of HEMA makes it an excellent adhesion-promoting monomer and increases the wetting of dentin, it significantly increases the bond strength and reduces microleakage. HEMA also forms hydrogen bonds within the microporosities of demineralized dentin, undergoes hygroscopic expansion after polymerization, mechanically clamps to the substrate, and bonds to the dentin surface with stronger bonds [37]. According to this study, the presence of HEMA resulted in less microleakage in the seventh-generation (Palfique bond) and eighth-generation (Palfique universal bond) adhesive system compared to the sixth-generation (One-up Bond F Plus) adhesive system [39]. In our study, the total-etch (OptiBond FL, Kerr) adhesive system showed lower microleakage than the self-etch (Gluma, Kulzer) adhesive system in all groups. Although both adhesive systems contain HEMA, the difference in HEMA content may be a factor for this result. Furthermore, the separate acid etching step in total-etch adhesive systems has a positive effect on better bonding of the restorative material to the tooth tissue.

Baltacıoğlu et al. (2024) investigated the effect of gravity on the microleakage po-tential of different bulk-fill composites with micro-CT device [40]. They found that gravity had no effect on microleakage but chemical composition of the composites sig-nificantly influenced the occurrence of higher microleakage ratios

In terms of microleakage formation, a significant difference was found in all groups. In the present study, Hypothesis 2, “The choice of restorative material does not influence microleakage formation” is rejected.

While there was no significant difference between the total-etch systems in terms of microvoid formation in all groups, only the G-aenial Posterior total-etch group had more microvoid formation than the G-aenial Posterior self-etch group. Accordingly, Hypothesis 3, “The choice of adhesive system (etch-and-rinse or self-etch) does not influence microvoid formation” is partially accepted. Depending on the adhesive system, a significant difference in microleakage was observed between the total-etch and self-etch systems in each restorative material group. According to these results, Hypothesis 4, “The choice of adhesive system (etch-and-rinse or self-etch) does not influence microleakage formation” was rejected. In vitro studies, including this one, have limitations. Notably, this study was conducted in a controlled laboratory setting. However, it is crucial to recognize that teeth and restorative materials experience ongoing stress and are influenced by various factors in real-world clinical conditions. Therefore, clinical conditions should be taken into account. Due to the different anatomical forms of the teeth, the ability to adjust the depth and width of the cavity in the same way is one of the limitations. Limitations include the fact that not all bulk-fill and short-fiber-reinforced resin composite materials and only two different adhesives were included in the study. However, it should be taken into consideration that different results can be obtained when bulk-fill and short-fiber-reinforced resin composite materials are applied with different adhesive systems.

Further studies under in vivo conditions using adhesive systems with different adhesive application procedures are needed for future research.

For Skyscan systems, the resolution is proportional to the diameter of the object; for a field of view of 10 mm, the resolution is typically 10 µm. This is a limitation of the methodology. Quantitative analysis based on a set of images with a resolution of 10 µm and above has a limitation; it does not detect small objects, of which there may be many. The study was conducted on first premolar teeth; this is a limitation. In order to examine the outcomes of posterior teeth, the study may also be performed on posterior teeth.

5. Conclusions

The use of adhesive systems and restorative materials according to the application procedures with appropriate instructions can prevent large microvoid potentials that may occur. MOD cavities showed less microleakage with three-step total-etch adhesive systems than with one-step self-etch adhesive systems. Within the limitations of this study, preliminary conclusions were drawn regarding the microvoid and microleakage potential of a microhybrid and two bulk-fill resin composites used in deep cavities when applied with different adhesive materials. In the present study, a significant difference was observed in terms of microvoid formation between the resin composites prepared with the incremental technique and the groups prepared with the bulk technique. The use of conventional composites alone may result in microvoid formation in cavities exceeding 4 mm. Microvoid formation in resin composites can also be attributed to the dentist’s skill and improper placement of bulk-fill composites in the cavity during the restoration process. Total-etch groups showed better results with less microleakage than self-etch groups. As a consequence, three-step total-etch adhesive systems can be preferred to prevent microleakage, especially in teeth with large and deep cavities. Within both the adhesive systems used in the study, Filtek One Bulk Fill Posterior showed the lowest leakage values followed by EverX Posterior and G-aenial Posterior. G-aenial Posterior showed the highest microleakage in both adhesive groups. These results indicate that bulk-fill resin composites can be used safely in deep cavities. However, it should be taken into consideration that different results can be obtained with the application of different adhesive systems.