Impact of Thermal Aging on Marginal Adaptation in Lithium Disilicate CAD/CAM Crowns with Deep Proximal Box Elevation

Introduction

Ceramic restoration computer-aided design (CAD) and computer-aided manufacturing (CAM) technology has advanced significantly to meet the growing need for aesthetically pleasing indirect ceramic restorations that can be completed in a single visit via chairside CAD/CAM systems, while maintaining high success rates [1–3]. However, the extensive loss of tooth structure, particularly in the proximal areas, presents a significant challenge when indirect restorations are needed. Teeth with deep proximal carious lesions or defective restorations can have little or no enamel available for successful bonding [4]. Additionally, subgingival margins cause the impression process to be more complex, leading to difficulties in reaching a dry environment for cementation and total complete elimination of additional cement [5].

Addressing these complexities requires careful clinical considerations [6]. For teeth with damage extending below the gingiva, surgical crown lengthening and orthodontic or surgical extrusion is generally advised to streamline the restoration [7]. These methods can often result in increased loss of attachment, exposure of the root surface, dentin hypersensitivity, and aesthetic issues and can potentially delay the placement of the final restoration [7]. An alternative and more conservative approach is deep proximal box elevation, also known as deep margin elevation (DME) [6]. The DME technique, compared with surgical crown lengthening, reduces effects of invasiveness on periodontal tissue and their inflammatory process [8].

DME is a restorative technique that offers significant benefits. Placing a restorative material as a base facilitates the extension of margins for crown preparation [5,9,10]. This procedure not only saves time, resources, and biological tissue but also simplifies the management of tooth-restoration intaglio surfaces. DME streamlines the process of making impressions, rubber dam isolation, and delivering bonded restorations. Importantly, DME ensures margin quality and strength that are comparable to those of restorations without DME, instilling confidence in its potential [5,7,8,11].

The proper selection of restorative materials for DME is crucial for ensuring the overall quality and longevity of the restoration and for maintaining periodontal health [8]. In deep carious lesions, resin-modified and highly viscous conventional glass ionomer filling and flowable and backable resins have shown variable long-term outcomes [12–15]. However, restorative materials can develop defects at the interfaces during placement or over time because of the burden of serious complications. Given that ceramic restorations are placed in cavities completely surrounded by enamel, their clinical prognosis is superior [16,17]. The main issue arises when trying to elevate deep proximal areas where crowns are bonded to a restorative material rather than to enamel surface. This issue highlights the need for caution and attention in material selection [17–19].

In 2021, advanced lithium disilicate glass-ceramic CEREC Tessera was introduced by combining 2 primary crystals in its blocks: virgilite crystal (Li0.5 Al0.5 Si2.5 O6), which is lithium aluminum silicate, and lithium disilicate (Li2 Si2 O5) [20–22]. This unique chemical structure provides many benefits, such as quick processing though rapid sintering, which can be achieved in as quickly as 4.5 min [21,23,24]. It is precisely intended for use in full and partial coverage and veneers [25] and consists of 90% lithium disilicate crystals and 5% virgilite content by volume [26].

The reduction in margin adaptation is related mainly to the factors of mechanical loading, cyclic stresses, and degradation of the restoration interface [27]. Marginal discrepancy is a crucial factor, as failing to meet the criteria for direct and indirect restoration can lead to failed restorations [28,29]. Microscopic analysis of the marginal adaptation of indirect restorations has been commonly used in the literature. Various microscopes have been used, such as optical microscopes or stereomicroscopes [30]. These methods allow quantification of the marginal gap in a noninvasive and less time-consuming manner. Scanning electron microscopy has also been used to obtain more detailed and specific images, because of its higher resolution [30–32].

It is always important to observe the performance of dental materials in an environment that mimics the thermal changes in the oral cavity [33]. The most common procedure for thermal or artificial aging is thermocycling, which is a system that simulates the natural aging process of dental materials by exposing them to the recurrent temperature range of 5 to 55°C in the mouth [34,35]. Thermal aging exposes the material to repeated cycles of temperature change, which simulates the effects of thermal stress on the material in the oral atmosphere. One year of clinical and oral services can be simulated by 10 000 thermal aging cycles [34,36,37].

In previous studies that assessed DME using different restorative materials, Zaruba et al [38] used a microhyprid composite, and Roggendorf et al [39] used a different brand of nanohyprid composite with different types of adhesive materials, while others used bulk-fill flowable and bioactive resin composite [40,41]. All studies concluded significantly different values of marginal adaptation, with an acceptable value.

In the present CAD/CAM study, we aimed to evaluate the effects of thermocycling on the DME relocation of subgingival cavity outlines in 80 molar teeth using advanced lithium disilicate ceramic materials.

The vertical marginal adaptations were measured between the advanced lithium disilicate CAD/CAM crown to the restorative material interface (cervical area) and the restorative material to the tooth interface (apical area). The null hypothesis in the present study is that no significant differences exist between the different restorative materials used for this technique. Additionally, no significant differences exist in the vertical marginal adaptation values (μm) between the advanced lithium disilicate CAD/CAM crown and the restorative material interface (cervical area) or between the restorative material and the sound tooth structure (apical area).

Material and Methods

STUDY DESIGN AND SAMPLE SIZE ESTIMATION:

This laboratory study was approved by the Ethics in Research Committee at Jazan University (REC-44/07/496). Healthy intact human mandibular molars, extracted for periodontal reasons, were used for this in vitro study after patient consent was signed. The teeth were inspected to ensure that no extensive wear, caries, cracks, fractures, or previous restorations were present. The teeth used presented comparable morphologies and dimensions of 8.4±0.35 mm, 11.37±0.28 mm, and 10.86±0.42 mm for the occlusocervical height and buccolingual and mesiodistal dimensions, respectively [32,42]. G*Power (version 3.1.9.7, 2020, Heinrich, Heine University Düsseldorf, Düsseldorf, Germany) was used to calculate the specimen sizes on the basis of a 5-group assessment. The alpha coefficient was set at 0.05 with a power of 95%, and an effect size of 0.40 was used to calculate the sample size.

CLASS II CAVITY PREPARATION:

A total sample of 80 teeth was used, and the teeth were equally divided into 5 groups. Prior to the cavity preparation procedure, teeth were perpendicularly fixed into a resin block up to a level of 3 mm apical to the cementoenamel junction. A class II slot cavity was designed for the proximal box, which was 4 mm in width, had a 1.5-mm axial depth, and had a cervical margin 2 mm below the cementoenamel junction in dentin [43]. The cavity outline and depth were checked with a periodontal probe to ensure standardization (Figure 1A, 1B).

TEETH GROUPING AND RESTORATIVE PROCEDURES:

The prepared teeth were distributed into 5 groups with 16 samples in each group. The samples were distributed into the control group (without cavity preparation and no restoration) 4 other groups that were divided equally according to the restorative materials. Four restorative materials were tested: a glass ionomer cement (Fuji IX, GC), bulk-fill flowable resin composite (SDR Bulk Fill Flow, Dentsply Sirona), bioactive resin composite (ACTIVA BioActive Restorative, Pulpdent), and nanohybrid resin composite (Filtek Z350 XT, 3M ESPE). All teeth were restored following the manufacturer’s instructions for bonding, restorations, and curing.

For the SDR Bulk Fill Flow, ACTIVA BioActive Restorative, and Filtek Z350 XT groups, teeth were conditioned with a solution of 37% phosphoric acid and Total Etch (Ivoclar AG) for 15 s, washed with a water jet for 30 s, and dried with a gentle stream of air, leaving a moist surface. Then, Scotchbond universal adhesive (3M ESPE) was applied in 2 consecutive layers and light cured for 10 s via a curing unit (Elipar, 3M ESPE), followed by the application of the restorations for SDR Bulk Fill Flow, ACTIVA BioActive, and Filtek Bulk Fill in single increments. The samples were polymerized for 20 s each on the occlusal, buccal, and lingulal surfaces. Filtek Z350 XT was applied incrementally in oblique layers, with each layer less than 2 mm thick, and polymerized for 20 s each on the occlusal, buccal, and lingual surfaces. In Fuji XI, prepared cavities were conditioned via a cavity conditioner (GC). The conditioner was applied to the cavities for 10 s, after which they were rinsed thoroughly with water and dried without desiccation. This step was followed by the application of Fuji IX, which was allowed to set chemically for 6 min. All teeth were finished and polished via the Sof-Lex Diamond Polishing System (3M ESPE) according to the manufacturer’s instructions under constant load (1±0.25 kg), distance, and equal number of grindings in a single direction, with 1 kit of disks for every 5 samples [44]. All the samples were subsequently stored in deionized water at 37°C before crown preparation.

CROWN PREPARATIONS AND CAD/CAM CONSTRUCTIONS:

All teeth with and without their fillings were further prepared by an experienced prosthodontist for advanced lithium silicate with virgilite full crowns: thickness of 1.5±0.25 mm, 1.0±0.25 mm in the axial wall, and a shoulder finish line with a width of 1.0±0.25 mm, with rounded angles, using a silicone index of an unprepared tooth to attain the necessary tooth reduction. The finish lines were 1.0 mm above the cementoenamel junction at the enamel surface for all groups, with 12° total occlusal convergence [45,46].

The prepared and restored filled teeth were subsequently optically scanned via an intraoral scanner (CEREC Primescan; Dentsply Sirona). All the crowns were fabricated with a CEREC primemill system (Dentsply Sirona) via software (CEREC SW, Version 5.2.9, Dentsply Sirona) to design the desired crown contours and occlusal relationships. The crowns were milled from a prefabricated block of advanced lithium silicate Cerec Tessera (Cerec Tessera A2, Medium Translucency, C14; Dentsply Sirona) at a standard milling speed. After recovering the crowns from the milling chamber, the crowns were trial-fitted to the preparation with a silicone material (Fit Checker, GC America, Alsip, IL, USA) to stabilize the crown and adjust it to ensure complete seating. After the final adjustments, the crowns were cleaned, dried, and glazed via universal stain and glaze liquid (Dentsply Sirona). Finally, according to the manufacturer’s instructions, the crowns were fired in a porcelain, in an oven under vacuum, to complete the crystallization processes [23].

CROWNS CEMENTATION:

For advanced lithium disilicate crown cementation, the fitting surfaces of the crowns were abraded with airborne 50-μm aluminum oxide particles (Polidental Indústria e Comércio, Cotia, SP) at 2.5 bar (6-mm distance) for 3 s, cleaned in an ultrasonic bath of 99% isopropanol for 3 min, and air-dried. Self-etching 3M Primer 94 (3M, St. Paul, MN, USA) was mixed, applied to the tooth and restorative material surface, scrubbed with a microbrush for 15 s, and gently air-dried. After the surface treatments, the advanced lithium disilicate crowns were cemented with self-adhesive resin cement (3M RelyX Unicem 2 Automix, St. Paul, MN, USA). For the cementing process (Figure 1C), the crown-to-cement teeth were loaded into the customized surveyor with an assigned load of 40 N during the cementation procedure, which simulated the pressure of the thumb, following a previously established protocol [47]. All the steps of advanced lithium disilicate cementation were conducted according to the manufacturer’s instructions. Finally, excess cement was removed from each crown-tooth-restorative material and lightly polymerized for 60 s via LED polymerization. The samples were stored in distilled water at 37°C for 72 h before thermocycling and vertical marginal adaptation measurements.

THERMOCYCLING OF THE CEMENTED SAMPLES:

The samples were subjected to 15 000 thermocycles by a thermocycling device (Thermocycler, SD Mechatronik, Feldkirchen-Westerham, Germany) in cold (5°C) and hot water (55°C) successively, the dwell time was 30 s, and vertical marginal adaptation measurements were assessed and performed before and after thermocycling [33–35,38].

VERTICAL MARGINAL ADAPTATION MEASUREMENTS:

Vertical marginal adaptation was assessed by measuring the vertical distance between the advanced lithium disilicate crown margin and the restorative material interface (cervical area) and the vertical marginal adaptation between the restorative material and the tooth structure interfaces (apical area). All the samples were sealed in the locking device and examined at ×50 magnification under the lens of a digital microscope (Digital Microscope, KH-7700, Hirox-USA, Inc, USA). The digital camera, light source, liquid-crystal display monitor, computer, and software were all integrated into this microscope. Vertical marginal gaps were measured for each sample in the area of class II restoration-crown-tooth and crown-tooth for the control group. Three points were measured at equidistant landmarks at the mesiobuccal, midpoint, and mesiolingual locations along the cervical circumference for each crown-restorative material interface, which was referred to as cervical vertical marginal adaptation. The same measurements were repeated at class II mesial side/restorative material-tooth interfaces at the mesiobuccal, midpoint, and mesiolingual positions and referred to as apical vertical marginal adaptations (Figure 2). Measurements were obtained at baseline and after thermocycling.

Each measurement was repeated 2 times, after which the average was calculated and recorded in micrometers (μm). The images were captured and automatically transferred to digital imaging software, and the lens was calibrated each time. The average value in micrometers was calculated via the measurement software available via the stereomicroscope and used to calculate the range values at the measurement points [48–50], as shown in Figure 3A and 3C at the cervical area and in Figure 3B and 3D at the apical area before and after thermocycling.

STATISTICAL ANALYSIS:

The Statistical Package for the Social Sciences (SPSS) version 26.0 was used for statistical analysis. The mean and standard deviation (±SD) were calculated to describe marginal adaptation before and after thermocycling. The dependent t test was used to compare the mean marginal adaptation before and after thermocycling. One-way ANOVA was used to compare the mean change after thermocycling among the mesiobuccal, midpoint, and mesiolingual methods for each group. The mean change after thermocycling was compared among the 5 groups for each landmark. A P value of <0.05 was considered to indicate statistical significance.

Results

Table 1 shows the mean values (±SD) of vertical marginal adaptation before and after thermocycling for all 5 groups. No change in vertical marginal adaptation was observed after thermocycling in the control group. While changes were observed after thermocycling, the highest values were observed for the glass ionomer filling, followed by the nanohybrid composite resin and bioactive composite resin. The lowest values were recorded for the bulk-fill flowable composite in the cervical and apical areas.

Figure 4 shows vertical marginal adaptation at the cervical and apical interfaces before and after thermocycling. At the cervical interface, the mean values (±SD) of vertical marginal adaptation at the mesiobuccal, midpoint, and mesiolingual landmarks before and after thermocycling in the glass ionomer filling material were 3.81 (±0.00), 3.81 (±0.00), and 3.81 (±0.00), and 12.39 (±0.98), 15.31 (±2.76), and 11.01 (±0.77), respectively; in the nanohybrid composite resin, 8.79 (±2.13), 7.15 (±0.85), and 7.37 (±1.06), and 3.88 (±1.39), 3.34 (±0.85), and 3.56 (±1.06), respectively; in the bioactive composite material, 3.81 (±0.00), 3.81 (±0.00), and 3.81 (±0.00), and 6.67 (±1.02), 7.60 (±1.52), and 6.67 (±1.02), respectively; and in the bulk-fill flowable composite, 3.81 (±0.00), 3.81 (±0.00), and 4.50 (±0.73), and 5.83 (±1.23), 6.25 (±0.96), and 5.73 (±0.02), respectively.

At the apical interface, the mean values (±SD) of the vertical adaptations at the mesiobuccal, midpoint, and mesiolingual landmarks before and after thermocycling in the glass ionomer filling material were 3.81 (±0.00), 3.81 (±0.00), and 3.81 (±0.00), and 13.95 (±6.22), 13.00 (±6.29), and 14.30 (±7.17), respectively; in the nanohybrid resin, 3.89 (±0.20), 3.81 (±0.00), and 3.81 (±0.00), and 8.19 (±1.19), 6.51 (±0.69), and 8.40 (±2.03), respectively; in the bioactive resin, 3.81 (±0.00), 3.81 (±0.00), and 3.81 (±0.00), and 6.59 (±1.08), 6.51 (±1.03), and 6.11 (±0.91), respectively; and in the bulk-fill flowable composite, 3.81 (±0.00), 3.81 (±0.00), and 3.91 (±0.18), and 6.19 (±0.89), 6.81 (±0.98), and 5.81 (±0.13), respectively.

Moreover, the mean (±SD) change after thermocycling at the mesiobuccal, midpoint, and mesiolingual landmarks in the cervical area was greatest for the glass ionomer filling: 8.58 (±0.98), 11.501 (±2.762), and 7.196 (±0.768), respectively, followed by the nanohybrid material: 3.88 (±1.39), 3.34 (±0.85), and 3.56 (±1.06), respectively; bioactive composite: 2.86 (±1.02), 3.79 (±1.52), and 2.86 (±1.02), respectively; and bulk-fill flowable material, 2.02 (±1.23), 2.44 (±0.96), and 1.23 (±0.73), respectively.

Similarly, the mean (±SD) change after thermocycling at the mesiobuccal, midpoint, and mesiolingual landmarks in the apical area was greatest for the glass ionomer filling material, 10.134 (±6.218), 9.191 (±6.294), and 10.484 (±7.166), respectively; followed by the nanohybrid resin, 4.31 (±1.19), 2.70 (±0.69), and 4.59 (±2.03), respectively; bioactive resin, 2.78 (±1.08), 2.69 (±1.03), and 2.30 (±0.91), respectively; and bulk-fill flowable resin, 2.38 (±0.89), 2.99 (±0.98), and 1.89 (±0.24), respectively.

The overall mean (±SD) change after thermocycling in the cervical and apical areas was greatest for the glass ionomer filling material, 9.091 (±1.147) and 9.936 (±6.376), respectively; followed by the nanohybrid resin composite material, 3.59 (±1.03) and 3.87 (±0.97); bioactive resin material, 3.17 (±0.81) and 2.59 (±0.21); and bulk fill flowable composite, 1.89 (±0.60) and 2.42 (±0.64). The dependent t test revealed significant differences in the scores of vertical marginal adaptations before and after thermocycling for all the materials at the mesiobuccal, midpoint, and mesiolingual landmarks in both the cervical and apical areas (Table 2).

ANOVA revealed significant differences in the change scores after thermocycling among the 3 different landmarks for the glass ionomer filling material (P<0.0001) and the bulk-fill flowable resin composite (P=0.004) in the cervical area. Moreover, significant differences in the change scores after thermocycling among the 3 landmarks were recorded for the bulk-fill flowable composite resin (P=0.001) and nanohybrid composite resin (P=0.001) in the apical area (Table 3). Also, ANOVA revealed significant differences in the change score of vertical marginal adaptation after thermocycling between all materials in the cervical area (all P<0.0001) and apical area (all P<0.0001) (Table 4).

Discussion

The aim of the present study was to evaluate the vertical marginal adaptation between different restorative materials used for DME and the newly developed advanced lithium disilicate CAD/CAM crowns before and after thermocycling. Regardless of the material used and the location at which the measurements were taken, the marginal adaptations between advanced lithium disilicate and the restorative material used for DME were within the clinically acceptable range (<120 μm). Thermocycling significantly affected the score of vertical marginal adaptation among all the materials in both the cervical and apical areas. The change in marginal adaptation after thermocycling was in the following descending order: glass ionomer filling material > nanohybrid resin composite material > bioactive resin composite material > bulk-fill flowable resin composite material, but all of these values were within the clinically acceptable range.

Thus, the null hypothesis was mainly accepted, as a significant difference was recorded between and within the used restorative materials, except at the cervical (P=0.051) and apical (P=0.370) areas for bioactive resin composite, nanohybrid resin composite at the cervical (P=0.391) area, and glass ionomer fillings (P=0.848) at the apical area. This is somewhat parallel with values documented in previous studies [48,51,52], although they used different systems for marginal discrepancy measurements. However, the values in the present study were smaller than those recorded earlier [10,11], which could be related to the newer restorative materials used in this study and the improvements of the composite component as filler type, the organic and in-organic components, and the use of advanced lithium disilicate ceramic crowns.

According to the recent literature and manufacturer information, advanced lithium disilicate offers numerous benefits, including quick crystallization, finishing in just 300 s, which speeds up the engineering progress [22]. In addition, increased aesthetics and mechanical properties are associated with advanced lithium disilicate, with respect to marginal fit and accuracy. A previous study by Canals revealed that advanced lithium disilicate produced very good marginal integrity and marginal gaps that were within clinically acceptable ranges (<120 μm) [51]. The mechanical properties of advanced lithium disilicate are affected by many factors, such as hydrothermal aging and surface finishing and polishing, and reveal promising potential for their long-term clinical performance [52]. Combining excellent optical and mechanical properties with time-efficient and predictable treatment options can be achieved via the use of CAD4AM dentistry and advanced lithium disilicate ceramics [21,23].

Compared with crown lengthening, DME combined with full-coverage crowns has been shown to increase survival [7,9,53]. The DME did not affect the fracture strength, fracture type, or reparability of CAD/CAM-manufactured lithium disilicate [21,48,54,55]. Also, the DME had no effect on either the marginal integrity or the fracture behavior of molars restored with feldspathic ceramic onlays and CAD/CAM-fabricated composite onlays [13,56].

The main issue with deep proximal boxes when bonding ceramic restorations is that moisture control can pose a challenge, as the placement of rubber dams can be difficult [5,9,29,57]. For this problem to be solved, DME can be performed with a restorative material to facilitate the placement of rubber dams and adhesive luting [10]. The bonding of confined areas of a proximal box to a resin composite is a relatively fast procedure and poses less risk for contamination than does luting multiple surfaces to indirect restorations [5,7,8,10].

One of the factors to consider during DME is the different restorative materials used. In this study, we selected 4 restorative materials: glass ionomer cement (Fuji IX, GC), a bulk-fill flowable resin composite (SDR Bulk Fill Flow, Dentsply Sirona), a bioactive resin composite (ACTIVA BioActive Restorative, Pulpdent), and a nanohybrid resin composite (Filtek Z350 XT, 3M ESPE), which are commonly used, have good biomechanical properties, and are easy to handle. All the materials produced excellent marginal adaptation levels that were far below the levels accepted for clinical application. A recent systematic review stated that lithium disilicate materials using CAD/CAM systems recorded an excellent material for vertical marginal adaption [58].

The quality of CAD/CAM restorations fabricated from lithium disilicate placed on aged resin-based composite restorations used for DME documented a good clinical behavior [13]. In a study by Zaruba et al, the marginal adaptation of ceramic inlays to either 1 layer or 2 layers of highly filled microhybrid composite was evaluated and compared with the adaptation of ceramic inlays to enamel or dentin without DME; the enamel showed the best gap-free margins after thermomechanical loading [38]. However, marginal adaptation in groups with DME filled with highly filled microhybrid composites in both layers was not significantly different from that in groups with DME bonded directly to dentin [38]. Another study, by Roggendorf et al, evaluated the effects of the different types of resin composite and the number of layers used in DME on marginal adaptation to a standardized mesial-occlusal-distal cavity [39]. Different types of self-adhesive resin cement or nanohybrid composite restorations with 1 or 3 layers were tested for their ability to adapt directly to dentin without DME [39]. No significant difference was found between bonding directly to dentin and proximal boxes elevated with 3 layers of nanohybrid composite restoration [11,12,32,39].

According to previous studies, glass ionomer cements are less irritating to periodontal tissues than are resin composites and can be used when contamination is a concern; however, they have poor mechanical properties, high solubility rates, and inadequate long-term bond strength to the tooth surface [59–61]. This situation may explain why the greatest change in vertical marginal adaptation in the present study occurred after thermocycling. Bulk Fill Flow and ACTIVA produced good marginal adaptation and exhibited the least change in vertical marginal adaptation after thermocycling. This result can be attributed to their composition. Bulk Fill Flow is a highly filled flowable composite and results in minimal volumetric shrinkage and polymerization stresses [38,40,55]. In addition, the material can flow easily along the cavity surface, ensuring optimal adaptation to any tooth irregularities [3,40,43,62]. ACTIVA is also a flowable material with lower filler loading and reduced polymerization shrinkage that can chemically adhere well to the tooth surface, leading to better marginal adaptation [40,63,64]. In a previous study, the marginal adaptation of CAD/CAM ceramic onlays to flowable or microhybrid composites used for DME was assessed. No significant difference in marginal quality was observed between the 2 composites, compared with bonding directly to dentin; however, bonding to enamel was significantly greater [5,65,66].

The use of microscopic analysis of the marginal adaptation of indirect restorations has been commonly reported in the literature. Various microscopes have been used, such as optical microscopes or stereomicroscopes [30]. These methods allow quantification of the marginal gap in a noninvasive and less time-consuming manner. Scanning electron microscopy has also been used to obtain more detailed and specific images because of its higher resolution [30–32,50]. Almost similar marginal adaptation values in microns were documented by using the similar system for measurements, but the authors tested a different CAD/CAM material from that used in the present study [67].

McLean and Fraunhofer established that 120 μm is the maximum limit that can be accepted clinically for marginal gaps [68]. None of the restorative materials used in the present study exceeded this limit, even after thermocycling at all sites cervically and apically. However, thermocycling did significantly change the marginal adaptation of all restorative materials. This increase in the marginal gap can be attributed to hydrolytic degradation and material deterioration caused by mechanical stresses [30]. After thermocycling for the restorative materials and advanced lithium disilicate crowns used for DME, similar vertical discrepancy values were recorded in the present study and previous studies [48,69].

In this study, we attempted to highlight some of the issues related to the marginal opening of crown-restoration and restoration-tooth interfaces following the restoration of deep proximal cavities. However, additional clinical studies are needed to be able to directly correlate the results of this study with actual clinical conditions. Also, we recommend a study with a higher number of thermocycling units and with dynamic monitoring of the marginal adaptions over a longer period.

Conclusions

Within the limitations of this microscopic study, the following conclusions can be drawn. The materials used for DME showed vertical marginal adaptation within the acceptable ring (up to 120 μm). Among the tested materials, the highest values for vertical marginal adaptation were recorded in the glass ionomer filling group, followed by the nanohybrid resin composite group and bioactive resin composite group, whereas the lowest values were recorded in the bulk-fill flowable resin composite group in the cervical and apical areas. The values of marginal adaptation in micrometers were almost the same at the 3 different landmarks, mesiobuccal, midpoint, and mesiolingual. Finally, thermal aging increased the values of the tested materials at both the cervical and apical areas.