2025 Volume 25 Issue 1 Pages 5-9
Purpose: The study aimed to compare the gap formation between dentin and resin composite using optical coherence tomography (OCT) and the leakage test.
Methods: Three cavities were prepared for each bovine tooth (coronal, cervical, and root). Cavities were filled by Scotchbond Universal Plus Adhesive and Filtek Supreme Flowable Restorative. After 24 h storage in water, the interface was analyzed at the bottom of the cavity using OCT. They were then cut and immersed in 50% ammoniacal silver nitrate (AgNO3) solution and photo-development solution for a leakage test. Next, energy-dispersive X-ray spectroscopy (EDS) assessed the cavity bottom interface. Two-dimensional (2D) images of both devices were analyzed to calculate the sealed interface percentage (SI%). The mean SI% values for adhesive were statistically analyzed by multiple comparisons using the Kruskal-Wallis test followed by Dunn's test with Bonferroni correction, with the significance level set at α = 0.05.
Results: No significant differences were observed among the three sites for either OCT (P < 0.05) or EDS (P < 0.05). Significant differences were observed between OCT and EDS at each site (P = 0.005, P < 0.001, P = 0.004, respectively). When comparing OCT and EDS, EDS detected significantly more white lines (lower SI%) in all locations.
Conclusion: The OCT device showed different detection results compared to the conventional EDS-based leakage method. On the other hand, there was no significant difference in the measurements across different sites of the tooth with either method.
Resin composite (RC) offers a significant benefit in that it is less invasive and preserves aesthetics. For RC restorations to last long-term, it's not just about achieving an attractive appearance; durability is equally crucial. A strong bonding interface is essential for the success of the resin composite. However, resin-based composite materials often shrink during polymerization, creating gaps at the interface between the resin and the tooth [1]. This interfacial microleakage may result in retention loss, secondary caries, or marginal discoloration [2]. The primary issue that affects the durability of dental restorations with RC is microleakage, which occurs due to gaps between the resin composite and the cavity wall [3].
An assessment of the sealing capacity is necessary to anticipate the therapeutic effectiveness of adhesives in the development of the occurrence of secondary caries [4]. Traditionally, dye-penetration leakage tests and microscopic examinations have been commonly used techniques to identify interfacial gaps in vitro [5]. Usually, these approaches require dividing the teeth into sections to evaluate the interface, which raises worries about potential damage to the sample. Leakage tests are the recommended method for evaluating the level of marginal adaptation or gap generation. Nevertheless, leakage tests are hindered by their labor-intensive nature, time-consuming process, and sensitivity to technique [6]. Furthermore, this method provides a restricted two-dimensional (2D) representation of the marker fluid distribution, making it difficult to accurately determine the deepest point of leakage [7]. Hence, the validity of microleakage assessment protocols remains a subject of controversy due to the diversity in methodologies employed worldwide [8,9].
Recently, many studies of nondestructive gap evaluation using optical coherence tomography (OCT) have been reported [10,11]. OCT employs low-coherence interferometry to obtain cross-sectional images without the need for invasive procedures. OCT has proven to be quite useful in visualizing the interior biological system at a very small scale, without the need for specimen cutting or processing. This technique enables immediate visualization of the microstructural details of tissues and biomaterials. Most of the time, a light beam, which is often a laser, is directed onto a sample. The intensity of the light that is scattered back from within the medium provides detailed information on the structure at different depths. The computer can rebuild a visual image based on the signal [12]. OCT has been identified as an effective instrument for analyzing the creation of gaps at the contact between a tooth and a dental restoration [13]. OCT was integrated with a low-coherence near-infrared light source. Near-infrared light is a high-sweeping laser that repetitively sweeps from 1,260 to 1,360 nm (centered at 1,310 nm) with a 20 kHz rate. The axial resolution of the system in air is 12 µm, corresponding to 7 µm within dental tissue with a refractive index of approximately 1.5 [14]. A handheld scanning probe attached to the OCT device was positioned 5 cm from the specimen, aligned perpendicularly. A total of 30 images were acquired. The comparison of resolution and imaging depth for ultrasound, OCT, and confocal microscopy. Standard clinical ultrasound can image deep structures. But the resolution is limited. Higher frequencies yield finer resolution, but ultrasonic attenuation is increased, and it limits the image penetration. The axial image resolution in OCT ranges from 1 to 15 um, and it is determined by the coherence length of the light source. In OCT, the observation area is limited to the penetration depth of light. In most biological tissues, the imaging depth is limited to 2-3 mm. Confocal microscopy has submicron resolution, but optical scattering limits the imaging depth to a few hundred micrometers in most tissues [15]. One image was captured per sample for this study. The image format was a BMP file (.bmp) and Microsoft Excel Comma Separated Values File (.csv).
Tooth structures can vary significantly based on their locations. The density of dentinal tubules is higher in the crown dentin than in the root dentin [16]. The dentinal tubules in the typical mid-coronal regions are wider and more numerous compared to those in the cervical region of the crown [17]. OCT is a device that uses light for measurements. The configuration of dentinal tubules, such as their direction, dentin thickness, and the presence or absence of enamel, may vary depending on the area. As a result, light transmittance may also differ. Thus, OCT may not always provide precise detections depending on the specific tooth region, prompting further exploration of this issue in this experiment. As a result, the study focused on comparing the difference between using OCT and leakage tests for evaluating the gap between RC and internal cavities. The first null hypothesis is “there is no difference in the performance of the detecting device” and the second is “there is no difference in the measurements across different sites of the tooth.”
A schematic image of sample preparation is shown in Fig. 1. For this study, 10 extracted and frozen bovine incisors (Tokyo Shibaura Zouki, Tokyo, Japan) were utilized. No animals were harmed during this study. They were placed in deionized water at a laboratory temperature of 23°C and unfreeze naturally. Tapered cavities were made with a diamond rotary cutting instrument (No.149, Shofu, Kyoto, Japan). The bur was connected to a water-cooled high-speed air turbine. The top diameter was 2.5 mm, the bottom diameter was 1.5 mm, and the depth was 2.0 mm for each cavity. Following the cavity preparation process as described above, the top diameter and depth of each cavity were measured to confirm that the standardized dimensions were achieved as intended. For each tooth, three cavities were prepared: one at the coronal site (CO), one at the cervical site (CE), and one at the root site (RO) [18]. The materials and their compositions used in this study are listed in Table 1. Scotchbond Universal Plus Adhesive (3M ESPE, St. Paul, MN, USA) was used in the cavities. On the surface of the dentin, the adhesive was used for 20 s, then the air was blown on it for 5 s and cured with light for 10 s. Subsequently, the cavities were filled by Filtek Supreme Flowable Restorative (A3, 3M ESPE). Then it was light-cured for 20 s [19]. A light-emitting diode (LED) light curing unit (VALO LED Light Curing Unit, Ultradent Products, South Jordan, UT, USA) was employed for light curing, with a high power mode (1,400 mW/cm2) [20]. All samples were kept in deionized water at 37°C for 24 h after bonding. Consecutive observation was performed on the same sample using optical coherence tomography (OCT) and energy-dispersive X-ray spectroscopy (EDS). Specimens were named into two groups. The 1st group observing OCT was named the “O” group. Following the OCT observation, the same specimen underwent an EDS test, with the second group designated as the “E” group.
| Brand name* | Composition | Lot number |
|---|---|---|
| Scotchbond Universal Plus Adhesive | Bis-GMA, 10-MDP, 2-HEMA, Vitrebond copolymer, ethanol, water, Adhesive initiators, fillers | 9106826 |
| Filtek Supreme Flowable Restorative (A3) | Bis-GMA, UDMA, TEGDMA, Bis-EMA, PEGDMA, silica, zirconia | 9299138 |
*Manufacturer 3M ESPE, St. Paul, MN, USA
The OCT (IVS-2000, Santec, Komaki, Japan) was employed in the present study. For the OCT examination, the specimens were stored in deionized water at 37°C for 24 h, taking care not to let the specimens dry out. The central part of the RC restoration was positioned for OCT scanning. After the scanning, these lines scanned by OCT were marked.
Leakage testA low-speed diamond saw (Isomet, Buehler, Lake Bluff, IL, USA) vertically sectioned all specimens along the previously recorded lines after OCT observation. After that, the sections were left in the dark for 24 h while immersed in a solution that included 50% ammoniacal silver nitrate (AgNO3). The specimens were then put in a photo-developing solution and left for 8 h under fluorescent light after a thorough rinsing with deionized water for 5 min. Afterward, the specimens underwent another rinsing process with deionized water for 5 min. Following this, samples were polished with SiC paper of grit sizes ranging from 600 to 1,000. Subsequently, they underwent a process of gold sputter coating. The morphology of the bonding interface was analyzed using scanning electron microscopy (SEM; JSM-IT100, Jeol, Tokyo, Japan) and energy-dispersive X-ray spectroscopy (EDS; JSM-IT100). The investigations were conducted at a working distance of 10 mm, with a magnification of 35× and an accelerating voltage of 20 kV [21,22].
Analysis of imagesThis study employed the Sealed Interface (SI%) area percentage criteria to analyze the images and measure the effectiveness of the adhesives [19]. The software (ImageJ version 1.53t, National Institute of Health, Bethesda, MD, USA) for analyzing images was utilized to import 2D OCT raw tomograms. To reduce background noise, a median filter was performed [5]. The analysis of the image was made easier by using an experimental technique for determining the threshold. This ImageJ Java plugin implemented this algorithm [23]. A rectangular region of interest (ROI) was chosen at the lowest portion of the restoration interface, omitting the specimen's surface. The ROI had a width of roughly 80 pixels, and the adhesive contact was positioned in the middle of the ROI. Each vertical line (A-scan) in the ROI represented a 2-pixel width, and the software plugin sorted the pixel values. The pixels with the highest 10% intensity values on each line were selected. Target pixels (white) were identified from the given set by comparing their intensity values to the sum of the background noise and median values. Only pixels with intensity values equal to or greater than this sum were considered target pixels. The remaining pixels were recognized as null, or black. Throughout the ROI, the plugin automatically calculated the gap or total percentage of white pixels. Subtracting the gap percentage from 100% yields SI%. Finally, the average of the SI% numbers for each sample was calculated.
From this study, the data was not normally distributed (Shapiro-Wilk test, P < 0.05) and did not display homogeneity of variance (Levene test, P < 0.05). Therefore, a non-parametric assessment was selected. Statistical analysis was conducted on the mean SI% values of each adhesive employing the Kruskal-Wallis test, followed by “Dunn's test with Bonferroni correction”. All statistical analyses were conducted using the SPSS ver. 26.0 (IBM, Armonk, NY, USA) with the statistical significance set to α = 0.05.
The analysis of the sealed interface percentage (SI%), median, and interquartile range values of OCT and EDS are summarized in Table 2. Representative OCT images are shown in Fig 2. A white line was noticed at the contact between the cavity wall and the RC (Fig. 2, indicated by finger pointers). Representative EDS images are shown in Fig 3. The signal of Ag was observed at the contact of the RC and the cavity wall (Fig. 3, indicated by finger pointers). For image analysis, SI% showed no significant difference among O-CO, O-CE, and O-RO and among E-CO, E-CE, and E-RO. Significant differences were found between ‘O-CO and E-CO’, (P = 0.005), ‘O-CE and E-CE’ (P < 0.001), and ‘O-RO and E-RO’ (P = 0.004). When comparing OCT and EDS, EDS detected significantly more white lines (lower SI%) in all locations.
| Site | OCT | EDS | ||||
|---|---|---|---|---|---|---|
| SI% | median | IQR | SI% | median | IQR | |
| Coronal | 93.2±3.4 aA | 93.3 | 5.7 | 67.3±10.0 bA | 69.3 | 16.4 |
| Cervical | 96.3±1.9 aA | 96.4 | 3.4 | 71.0±11.1 bA | 69.1 | 18.8 |
| Root | 96.1±3.0 aA | 97.3 | 4.2 | 76.5±9.7 bA | 79.2 | 13.5 |
The statistical measures of the SI% are presented, including the means and standard deviations. Values sharing the same lowercase letters within the same row and observation site did not exhibit significantly different behavior (P > 0.05). Values that have the same uppercase letters within the same column also did not exhibit significantly different behavior (P > 0.05).
R for resin composite, D for dentin, and E for enamel. Representative OCT image of (a) The coronal site group (O-CO). (b) The cervical site group (O-CE). (c) The root site group (O-RO). The white pointer shows the white line sites which means the presence of a gap.
R is resin composite. Representative EDS image of (a) The coronal site group (E-CO). (b) The cervical site group (E-CE). (c) The root site group (E-RO). The white pointer indicates the presence of silver ions, which were detected using EDS at the exact location on the tooth.
There are concerns about the adverse effects of tooth structural differences among sites because OCT is a light-based diagnostic instrument. For this reason, in this study, experiments were conducted by forming cavities in three sites: the crown, which contains enamel; the root, which does not contain enamel; and the cervical site, which is the boundary between the two sites. This study partially rejected the null hypothesis that the first null hypothesis is “there is no difference in the performance of the detecting device” and the second is “there is no difference in the measurements across different sites of the tooth”. The data of this study did not show statistically significant differences by sites. Therefore, OCT is considered to be an effective instrument regardless of the site. Although no significant statistical difference was observed, there was a noticeable trend for the results to differ depending on the location. Previous research has indicated variations in the number of dentin tubules across various locations [24], and that dentin adhesion is affected by the number of dentin tubules [25]. This may have influenced the variation in values. Several studies have found that the interface between the occlusal or mid-coronal tooth site and RC is more resistant to microleakage in vitro compared to the interface between the cervical tooth site and RC [26,27]. Besides the variation of dentin tubules in the crown, the structural differences between the enamel in the crown and cementum in the root should affect the microleakage level. However, in this study, the EDS test did not reveal any significant differences across the different sites of the tooth.
Significant differences were observed in detection values between OCT and EDS at all sites. Unlike OCT, which can observe samples in wet conditions, EDS is performed in a vacuum environment and, therefore, measures dried samples. In contrast to enamel, dentin is known for its high water content [28]. Sano et al. discovered that the silver nitrate tracer penetrated the hybrid layer of the dentin bonding interface without causing any gaps [29]. It was suggested that silver-filled nanoscale gaps surround exposed collagen fibrils [30]. These gaps may serve as possible locations for the hydrolytic breakdown of resin/dentin bonds [31]. The silver nitrate tracer demonstrates the ability to penetrate low-quality interfaces, even in the absence of detectable gaps. Its key advantage lies in its capacity to identify these compromised regions. In contrast, OCT exhibited lower accuracy compared to the leakage test, detecting fewer white areas. This indicates that, although OCT is proficient at identifying gaps, it is less effective in detecting more subtle interface defects.
The universal adhesive has the advantage of selective etching. However, the use of phosphoric acid etching on dentin is controversial. In this study, the area of the root without enamel was also measured. For that reason, phosphoric acid etching was not performed, and only the cavity floor was evaluated. Based on these findings, future research focusing on cavity walls, including enamel, is necessary. In general, flowable resin RC has been reported to improve wettability and conformability to cavity surfaces and walls due to their lower viscosity [32,33]. Another study showed that the gap size of high-viscosity resins was significantly larger than that of low-viscosity resins. This difference could be attributed to the excellent fluidity of low-viscosity resins, enabling them to better adapt to the fine structure at the edge of the cavity [34]. As a result, it has been reported that the fit of the cavity walls has improved [35,36]. In contrast to micro hybrids and nanohybrids, which consist of particles with diverse sizes and shapes created through grinding processes, the nanoparticles in flowable resin composites (RC) are precisely engineered from particles smaller than 100 nm.
Bakhsh et al. reported that the average gap percentage decreased when the curing time was extended from 10 s. Sometimes clinicians may also extend the light curing time in cases where they are concerned about the risk of gaps. Therefore, this study extended the light curing time based on the previous study by Kibe et al. [19].
This study utilized two different methods, each with its own advantages and disadvantages. Leakage tests are the preferred method for assessing the quality of marginal adaptation and the potential for gap formation. Although, many of these methods are invasive, preventing the reuse of specimens for re-evaluating leakage with different techniques, would be beneficial for validating previous results. Additionally, leakage tests can be difficult to standardize, making it challenging to obtain consistent, reproducible, and comparable results [37].
Previous research has demonstrated that OCT enables the immediate and on-site observation of tissue microstructure, eliminating the need for specimen removal, processing, or exposure to radiation [38]. OCT imaging has a significant advantage over radiography in that it may effectively decrease the amount of radiation exposure that is often linked with visual testing methods for dentistry. This characteristic makes OCT a safer alternative, particularly for sensitive patient groups, such as pregnant women, debilitated individuals, and young children, who may be more vulnerable to the adverse effects of radiation exposure. This study suggested that if a patient with these characteristics complains of hypersensitivity symptoms or other symptoms around the restoration, its suitability can be examined noninvasively without radiographs.
A limitation of this study is that only one flowable resin composite was used, meaning other resins may exhibit different trends. OCT devices in dentistry are capable of capturing images to a depth of 2-3 mm [39]. The 2-3 mm depth is adequate for examining superficial structures, such as enamel, the dentin-enamel junction (DEJ), and shallow caries. However, it limits the ability to assess deeper dental tissues, such as deep dentin lesions, pulp involvement, or large structural defects extending beyond the imaging range. OCT is less effective for evaluating deep interfacial gaps or marginal leakage in restorations if these defects are located beyond its penetration depth. The ability of light to pass through dental hard tissues and restorative materials in OCT imaging is limited by their refractive indices. In the future, further research is required to conduct OCT assessment in cavities that are located at greater depths and in closer proximity to the pulp compared to the scope of this study. The OCT device showed different detection results compared to the conventional EDS-based leakage method. On the other hand, there was no significant difference in the measurements across different sites of the tooth with either method.
Bis-EMA: ethoxylated bisphenol-A dimethacrylate; Bis-GMA: bisphenol A diglycidylether methacrylate; CE: cervical; CO: coronal; EDS: energy-dispersive X-ray spectroscopy; HEMA: 2-hydroxyethyl methacrylate; LED: light-emitting diode; MDP: 10-methacryloyloxydecyl dihydrogen phosphate; OCT: optical coherence tomography; PEGDMA: polyethyleneglycol dimethacrylate; RC: resin composite; RO: root; ROI: region of interest; SEM: scanning electron microscopy; SI%: sealed interface percentage; TEGDMA: triethyleneglycol dimethacrylate; UDMA: urethane dimethacrylate; 2D: two-dimensional
Not applicable
Not applicable
This work was supported by JSPS KAKENHI Grant Number, JP20K18524.
AHMK: conceptualization, investigation, methodology, data curation, formal analysis, visualization, and writing; TS: conceptualization, methodology, formal analysis, writing, review, editing, and supervision; TT: formal analysis, visualization, editing; MZ: writing, review, editing; KK: formal analysis, visualization, review; MI: methodology, formal analysis, review, editing; YS: conceptualization, methodology, formal analysis, writing, review, editing, and supervision. All authors read and approved the final version of the manuscript.
2) TS*: t.sato.ope@tmd.ac.jp, https://orcid.org/0000-0001-5515-7345
1) TT: tabope@tmd.ac.jp, https://orcid.org/0000-0001-9423-2424
1) MZ: zhaoope@tmd.ac.jp, https://orcid.org/0009-0003-2217-3762
1) KK: kibeope@tmd.ac.jp, https://orcid.org/0009-0005-6784-8414
2) MI: ikeda.csoe@tmd.ac.jp, https://orcid.org/0000-0003-2214-4980
1) YS: shimada.ope@tmd.ac.jp, https://orcid.org/0000-0002-0751-4819
The authors would like to thank the faculty and staff of the Institute of Science Tokyo for their guidance on experimental methodology.
All data generated or analyzed during the current study are available from the corresponding author upon reasonable request.
