Effects of adhesive application timing on the push-out bond strength of four calcium silicate-based cements: a comparative study in vitro

Abstract

Background

Information about recently introduced calcium silicate-based materials and their bond performance to coronal dentin after adhesive exposure is limited. The aim of this study was to determine whether dentin adhesive applied 1 h or 3 days after placing calcium silicate-based materials adversely affects their push-out bond strength (POBS) to dentin.

Methods

TheraCal PT (TC) and three premixed-type calcium silicate-based cements (CSBCs) – Endocem MTA Premixed Heavy (EC), One-Fil Putty (OF), and BC Universal RRM (RM) – were tested. Two coronal dentin discs with a thickness of 1 mm were obtained from each of 23 extracted human molars. The final total was 45 discs after excluding one that was damaged during preparation. Four cylindrical cavities were drilled in each dentin disc. The cavities of all dentin discs were filled with four calcium silicate-based materials, then grouped according to dentin adhesive application timing as early, delayed, or no adhesive (control). Prime & Bond Universal dentin adhesive was applied above the calcium silicate-based material after 1 h (early) and 3 days (delayed), respectively. After 7 days, POBS was measured with a universal testing machine. Five additional specimens were evaluated by scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS). The data were analyzed using Kruskal–Wallis tests with Bonferroni post-hoc correction (α = 0.05).

Results

Adhesive application at 1 h after premixed CSBC placement significantly reduced the POBS of EC compared with the control (P < 0.05). The POBS of TC remained high regardless of adhesive application. The SEM-EDS findings revealed dense polygonal crystalline structures in the control, whereas the delayed group had a mixture of these crystals and the adhesive layer. However, the adhesive penetrated the premixed-type CBSC, leaving visible filler particles on the surface of the early groups.

Conclusions

Within the limitations of this experimental design, adhesive application 1 h after placing premixed CSBC adversely affected the 7-day POBS of EC. The POBS was higher for TheraCal PT than premixed-type CSBCs when evaluated using Prime & Bond Universal in self-etch mode after storage for 7 days.

Background

Currently, various types of bioceramic materials are used in endodontics and restorative dentistry, particularly as pulp capping agents in vital pulp therapy (VPT) and regenerative endodontic procedures, due to their biocompatibility, sealing ability, and capacity to promote hard tissue formation [1, 2]. Among these materials, mineral trioxide aggregate (MTA), particularly ProRoot MTA (Dentsply Tulsa Dental, Johnson City, TN, USA), has been one of the most widely used in surgical endodontic treatment, perforation repair and VPT [3, 4]. Despite these advantages, ProRoot MTA has several drawbacks, including difficult handling characteristics, prolonged setting time, and potential for tooth discoloration [5, 6]. In particular, its extended setting time and hydration-dependent setting reaction often necessitate multiple clinical appointments [7].

Recently, premixed-type calcium silicate-based cements (CSBCs), such as Endocem MTA Premixed (Maruchi, Wonju, South Korea), One-Fil Putty (Mediclus, Cheongju, South Korea) and BC Universal RRM (Vericom, Chuncheon, South Korea), have been introduced. They do not require a mixing procedure and set rapidly in the presence of moisture and have an initial setting time of less than 10 min, and reduced discoloration risk while retaining the biologic advantages of MTA [8–10].

TheraCal PT (Bisco, Schaumburg, IL, USA), a dual-cured resin-modified calcium silicate, was introduced as a successor to TheraCal LC (Bisco, Schaumburg, IL, USA) [11]. TheraCal PT, which exhibited greater biocompatibility than TheraCal LC [12, 13], is utilized as a pulp capping agent in direct pulp capping and pulpotomy, as well as a protective liner [14, 15].

Achieving a durable coronal seal remains a critical determinant of the clinical success of VPT, as microleakage-induced bacterial ingress creates a conductive environment for pathogens, triggering a cascade of clinical complications that jeopardize the integrity of the pulp [16]. During pulpotomy, resin-modified glass ionomer (RMGI) can be applied onto CSBCs, followed by dentin bonding agent and composite resin without waiting for initial setting of CSBCs [17]. Conversely, direct pulp capping–particularly in cases of complicated crown fractures–involves a restricted exposure site where the application of RMGI within the prepared cavity might be precluded. In such clinical circumstances, dentin adhesives are simultaneously applied to the prepared tooth structure and CSBC surface, followed by composite resin restoration [18]. Dentin adhesives, with or without phosphoric acid etching, expose CSBCs to acidic environments. Previous studies have reported that acidic conditions may interfere with hydration and setting reactions of CSBCs, potentially compromising bond strength [19, 20]. In addition, acidic exposure has been associated with adverse changes in the physical properties and microstructure of CSBCs such as Biodentine (Septodont, Saint-Maur-des-Fossés Cedex, France) and EndoSequence Root Repair Material Putty (Brasseler USA, Savannah, GA, USA), including reduced microhardness and less crystalline structures [21].

The availability of fast-setting CSBCs has enabled single-visit VPT procedures. Previous studies reported that Biodentine and TotalFill BC Putty allowed immediate restorative procedures (12 min) after placement of pulp-capping biomaterials [22, 23]. In contrast, another study showed that the delayed restoration (7 days) of the NeoMTA2 (NuSmile, Houston, TX, USA) and NeoPutty (NuSmile, Houston, TX, USA) resulted in higher shear bond strength (SBS) values than immediate restoration (15 min) [24].

Although previous studies have evaluated the bond strength of selected calcium silicate-based materials under various restorative timings, information about recently introduced CSBCs and their bond performance to coronal dentin after adhesive exposure is limited. Therefore, we aimed to evaluate the push-out bond strength (POBS) of TheraCal PT and three kinds of premixed-type CSBCs to dentin according to dentin adhesive application timing (early, delayed, or without adhesive).

The null hypothesis was that the push-out bond strength of CSBCs to dentin would not be affected by the timing of dentin adhesive application or the type of calcium silicate-based cement.

Materials and methods

Sample size calculation

Effect size was set to 0.4 based on the results of a pilot study. Using G Power software (version 3.1.9.2, Heinrich-Heine-Universität) with the F test family and the ANOVA: Fixed effects, special, main effects, and interactions option (α = 0.05, power = 0.95), the required sample sizes were calculated to be 12 per group. Fifteen dentin discs for each group (early, delayed, control) were used.

Materials

The POBS of TheraCal PT (TC; Bisco, Schaumburg, IL, USA), Endocem MTA Premixed Heavy (EC; Maruchi, Wonju, South Korea), One-Fil Putty (OF; Mediclus, Cheongju, South Korea), and BC Universal RRM (RM; Vericom, Chuncheon, South Korea) was evaluated. EC, OF, and RM are premixed-type CSBCs, while TC is a dual-curing resin-modified calcium silicate-based material. The effects of dentin adhesive application were assessed using a universal dentin adhesive (Prime & Bond Universal; Dentsply Sirona, Charlotte, NC, USA). Table 1 shows the composition of these materials.

Table 1.

Composition of TheraCal PT, three premixed-type CSBCs, and a universal dentin adhesive

Product	Manufacturer	Composition	Initial setting time*	Setting mechanism
TheraCal PT (TC)	Bisco, Schaumburg, IL, USA	Silicate glass-mixed cement, polyethylene glycol dimethacrylate, Bis-GMA, barium zirconate, ytterbium fluoride, initiator	Light cure for 10 s	Dual cure
Endocem MTA Premixed Heavy (EC)	Maruchi, Wonju, South Korea	Zirconium dioxide, calcium silicate, calcium aluminate, calcium sulfate, dimethyl sulfoxide, thickening agent	3–5 min	Hydration
One-Fil Putty (OF)	Mediclus, Cheongju, South Korea	Tricalcium silicate compounds, zirconium oxide, hydrophilic polymer (thickening agents)	2–3 min	Hydration
BC Universal RRM (RM)	Vericom, Chuncheon, South Korea	Calcium aluminosilicate compound, zirconium oxide, thickening agent	< 10 min	Hydration
Prime & Bond Universal	Dentsply Sirona, Charlotte, NC, USA	PENTA, 10-MDP, Active Guard Technology crosslinker, photoinitiator, stabilizer, isopropanol, water		Light cure

*Initial setting time data were provided by the manufacturer

10-MDP 10-methacryloyloxydecyl dihydrogen phosphate, Bis-GMA Bisphenol-A-glycidyl dimethacrylate, PENTA dipentaerythritol penta-acrylate phosphate

Push-out bond strength test

This study was approved by the Institutional Review Board of Kyung Hee University Dental Hospital (KH-DT24019). Twenty-three human molars extracted due to periodontal disease were used. Teeth with dental caries, previous restorations, and fractures were excluded. Samples were stored in 0.1% thymol solution at 4 °C until use. Two dentin disc specimens (1 mm thick) were obtained from each tooth using a low-speed precision saw (ISOMET 5000; Buehler Ltd., Lake Bluff, IL, USA) with a diamond blade under continuous irrigation, yielding a total of 45 coronal dentin discs, as one disc was excluded due to specimen damage during preparation (Fig. 1A).

Fig. 1.

Schematic illustration of the dentin disc preparation. A two dentin discs obtained from the crown of each tooth, B a prepared cylindrical cavity with a diameter of 1.2 mm, C four cylindrical cavities prepared to be filled with four different calcium silicate-based cements

Four cylindrical cavities were prepared in each coronal dentin disc to simulate confined coronal dentin cavities using a cylindrical carbide bur (Mani #1958 long; Mani, Inc., Utsunomiya, Japan) with a diameter of 1.2 mm, positioned perpendicular to the disc surface and parallel to the original long axis of the tooth (Fig. 1B). A minimum distance of 1 mm between the cavity and the dentino-enamel junction was maintained to prevent sample fractures. Likewise, the distance between the cavities was > 2 mm (Fig. 1C). The dentin discs were immersed in 10 mL of 3% sodium hypochlorite for 10 min, followed by 10 mL of 17% ethylenediaminetetraacetic acid for 1 min to remove the smear layer [25, 26]. The specimens were immersed again in 10 mL of 3% sodium hypochlorite for 1 min, followed by 10 mL of distilled water for 1 min.

Each of the four pulp capping materials TC, EC, OF, or RM were applied to the four cavities in each dentin disc to minimize the effect of inter-tooth variability in dentin substrate properties, such as density and diameter of dentinal tubules, and elastic modulus. TC and EC were directly injected into the cavities using proprietary syringe tips provided by the respective manufacturers. TC was subsequently light-cured for 20 s using a light-emitting diode curing unit (Bluephase PowerCure, Ivoclar Vivadent, Schaan, Liechtenstein) with a wavelength range of 385–515 nm, an irradiance of 1200 mW/cm², and an energy density of 24 J/cm². OF and RM were inserted into the cavities using a resin applicator. All specimens were coded and covered with distilled water-soaked gauze to prevent dehydration.

The specimens were assigned to three groups (n = 15 each) using a computer-generated random number table (Microsoft Excel, Microsoft Corp., Redmond, WA, USA) to ensure simple randomization.

For the early and delayed groups, dentin adhesive was applied above the cement surface in self-etch mode after 1 h (early) and 3 days (delayed), respectively. The adhesive was applied for 10 s, followed by gentle air-drying for 10 s to evaporate the solvent. Subsequently, light-curing was conducted at irradiation distance of 2 mm using Bluephase PowerCure at an irradiance of 1200 mW/cm² for 10 s, resulting in an energy density of 12 J/cm². This procedure was designed to simulate the clinical situation during VPT, in which a dentin adhesive is applied over both the dentin and the CSBC during the final restoration. After bonding was complete, all specimens were stored at 37 °C under 95% relative humidity for one week. Humidity was maintained using a distilled water-fed humidifier. Controls without dentin adhesive were stored under the same conditions for 1 week, after which POBS was tested. Figure 2 shows the handling of dentin specimens in each group for POBS tests.

Fig. 2.

Schematic diagram of experimental groups according to dentin adhesive application timing

Specimens were mounted on a universal testing machine (AGS-X; Shimadzu, Kyoto, Japan). For the POBS test, a stainless-steel cylindrical pin with a diameter of 0.8 mm was driven in the apico-coronal direction at a crosshead speed of 1 mm/min until dislodgement. The maximum load at failure (F) was determined for each specimen. The POBS (MPa) was calculated by dividing the load at failure (F) by the bonded area (mm²) calculated as 2πrh, as described elsewhere [27, 28]:

, where r is the radius of the cavity in the dentin disc (mm) and h is the thickness of the dentin disc (mm).

Fractured surfaces were examined after the POBS tests, using a stereomicroscope at ×30 magnification. Failure modes were classified as adhesive at the dentin-CSBC interface, cohesive within CSBC, and mixed (two modes of failure combined).

Scanning Electron Microscopy (SEM) and Energy-Dispersive Spectroscopy (EDS) analysis

For SEM-EDS analysis, five additional dentin discs per group were prepared in the same manner as those for the POBS tests. SEM specimens were dried and sputter-coated with chromium. Images were magnified at ×5,000 and ×50,000 to evaluate the influence of dentin adhesive application on setting of CSBC. To examine the interface between the cavity wall and the CSBCs, as well as to evaluate the subsurface infiltration of CSBCs or their hydration products, dentin discs were fractured perpendicular to the outer surface and parallel to the long axis of the tooth. Examinations were performed at a magnification of ×20,000. SEM analysis was performed using Teneo VS (Thermo Fisher Scientific Inc., Hillsboro, OR, USA) and S-4700 (Hitachi High Technologies Co., Tokyo, Japan) Field Emission SEMs operated at an accelerating voltage of 10 to 15 kV, under high-vacuum conditions with a secondary electron detector.

The elemental distribution of the hydration products within dentinal tubules as well as the surface of CSBCs was analyzed using an EDS coupled to the SEM at ×5,000 magnification. The atomic percentage of calcium (Ca at%) was measured at two randomly selected points for each subgroup (n = 10).

Statistical analysis

The normality of data distribution was evaluated using Shapiro–Wilk tests, and the homogeneity of variances was assessed using Levene’s tests. Since the assumptions for parametric analysis were not fully satisfied, non-parametric tests were applied. The effects of dentin adhesive application timing on POBS were evaluated using Kruskal–Wallis tests, followed by Bonferroni post-hoc correction. Comparisons of POBS among different CSBCs were conducted using Kruskal–Wallis tests with Bonferroni post-hoc correction. The Ca at% on the surface of CSBCs and in the hydration products within the dentinal tubules according to different adhesive application timings were analyzed for each CSBC material using Kruskal–Wallis tests and Bonferroni post-hoc correction. Correlations between the Ca at% and POBS were assessed using Spearman’s rank correlation analysis. All data were statistically analyzed using IBM SPSS Statistics 25.0 (IBM Corp., Armonk, NY, USA). The level of significance was set at α = 0.05.

Results

Push-out bond strength

The normality of the data was assessed using the Shapiro-Wilk test. While TC, EC, and OF groups followed a normal distribution (P > 0.05), RM group did not (P < 0.05). Levene’s test was conducted to assess homogeneity of variances; the results indicated that the assumption of homogeneity of variances was violated (P = 0.002). Therefore, the data were analyzed using non-parametric tests. Table 2; Fig. 3 show the mean POBS values of CSBCs to dentin under various adhesive application timing conditions. The early groups had lower POBS values compared with the control and delayed groups. However, only EC significantly differed (P < 0.05).

Table 2.

Push-out bond strength of four calcium silicate-based cements to dentin (MPa)

Adhesive application timing	TC	EC	OF	RM
Control	25.05 ± 8.85aA	17.45 ± 3.84aAB	14.92 ± 4.92aB	17.11 ± 6.35aAB
Early	18.05 ± 6.31aA	10.61 ± 3.89bB	12.49 ± 3.61aAB	12.74 ± 2.60aAB
Delayed	20.95 ± 5.92aA	16.64 ± 4.39aAB	15.35 ± 4.63aAB	14.38 ± 4.55aB

Different superscript capital letters in each row indicate significant differences among CSBCs (P < 0.05)

Different superscript lowercase letters in each column indicate significant differences among three groups (control, early, delayed) according to adhesive application timing of CSBCs (P < 0.05)

Fig. 3.

Box plots showing the distribution of push-out bond strength values in the test groups. Within each adhesive application timing group, experimental groups sharing the same superscript letter do not show statistically significant differences (P > 0.05). An asterisk (*) indicates a significant difference between adhesive application timing conditions for EC (P < 0.05). Boxes represent the interquartile range, the central line indicates the median, and whiskers represent the minimum and maximum values

Regardless of adhesive application timing, the highest POBS values were consistently found in TC among the tested materials. The POBS values were significantly higher for TC than EC in the early group and for TC than RM in the delayed group (P < 0.05 for both). The POBS was significantly higher for TC than OF in the control group (P < 0.05).

Based on the observed effect size, the results revealed strong main effects for both ‘material’ (η_p² = 0.249, P < 0.001) and ‘adhesive application timing’ (η_p² = 0.152, P < 0.001) factors.

Figure 4 shows the distribution of failure modes in specimens after POBS tests. Adhesive failure was the most prevalent mode for RM and OF in the early and delayed groups, whereas cohesive failure is more prevalent for EC in those groups. Cohesive failure is predominated in the TC regardless of adhesive application timing.

Fig. 4.

Distribution of failure modes according to adhesive application timing and CSBC type

SEM-EDS analysis

Figures 5, 6, 7 and 8 show SEM images of the surface morphology of CSBCs when adhesive was applied at 1 h (early groups) or 3 days (delayed groups) after CSBC placement or not (controls), then stored for 7 days under 95% relative humidity. The TC without adhesive had a mixed population of spherical and polygonal crystals (Fig. 5A, D, G). Adhesive infiltration of crystalline structures was evident at 1 h after TC placement (Fig. 5B, E, H), and dense crystalline structures were covered by adhesive at 3 days after placement (Fig. 5C, F, and I).

Fig. 5.

Representative SEM images of TC according to different adhesive application timing. A, D, G Control; B, E, H Early; C, F, I Delayed. Scale bars represent 50 μm (A–C) and 5 μm (D–I)

Fig. 6.

Representative SEM images of EC according to different adhesive application timing. A, D, G Control; B, E, H Early; C, F, I Delayed. Scale bars represent 30 μm (A–C) and 3 μm (D–I)

Fig. 7.

Representative SEM images of OF according to different adhesive application timing. A, D, G Control; B, E, H Early; C, F, I Delayed. Scale bars represent 30 μm (A–C) and 3 μm (D–I)

Fig. 8.

Representative SEM images of RM according to different adhesive application timing. A, D, G Control; B, E, H Early; C, F, I Delayed. Scale bars represent 30 μm (A–C) and 3 μm (D–I)

Polygonal crystals of various sizes are evident in the EC, OF, and RM without adhesive application (Figs. 6A, D and G, 7A, D and G and 8A, D and G). When the adhesive was applied 1 h after filling the discs with CSBCs, the EC, ,OF and RM surfaces were covered by an adhesive layer, leaving adhesive filler particles on the surface of CSBCs (Figs. 6B, E and H, 7B, E and H and 8B and E), while crystalline structure formation was also observed (Figs. 6H and 8H). In the specimens where the adhesive was applied 3 days following CSBC filling, the adhesive and crystalline structures were intermingled in the EC, OF, and RM (Figs. 6C, 7C and 8C). Higher magnification revealed crystalline structures on the surface of EC (Fig. 6F, I), OF (Fig. 7I), and RM (Fig. 8I). Calcium silicate hydrate gel has also formed on the surface of OF (Fig. 7F). The RM was densely coated with adhesive (Fig. 8F).

Figure 9 shows subsurface dentinal tubules facing the CSBCs. Intratubular crystals were examined in all groups. The intratubular crystals were smaller in the early (Fig. 9B, E, H, K) compared with the delayed (Fig. 9C, F, I, L) and control (Fig. 9A, D, G, J) groups for all materials.

Fig. 9.

Representative SEM images of the fractured surfaces showing subsurface dentinal tubules along the cavity wall. Control, Early, and Delayed groups of (A–C) TC, (D–F) EC, (G–I) OF, and (J–L) RM, respectively. Yellow arrows indicate hydration products of CSBCs within the dentinal tubules. Scale bar in A = 5 μm (applies to all panels)

All CSBCs had lower Ca at% on the surface in the early group, compared with those of the control and delayed groups (Table 3). The surface Ca at% of the TC in the delayed group was lower than that of the control (Table 3). The relationship between the surface Ca at% and POBS showed a weak positive correlation (P = 0.035, ρ = 0.24).

Table 3.

Calcium at% of CSBC surface and within subsurface dentinal tubules according to adhesive application timing

	Adhesive application timing	TC	EC	OF	RM
Surface of CSBCs	Control	19.7 (14.2 ˗ 23.9)a	12.7 (6.2 ˗ 16.5)a	10.2 (8.1 ˗ 17.9)a	8.4 (3.1 ˗ 12.6)a
	Early	0.1 (0 ˗ 1)b	0.8 (0.3 ˗ 2.2)b	1.2 (0.7 ˗ 3.5)b	0.4 (0.3 ˗ 1.9)b
	Delayed	0.3 (0.3 ˗ 0.7)c	2.4 (0.9 ˗ 9.8)a	9.9 (2.5 ˗ 13.4)a	2.5 (0.8 ˗ 3.8)a
Subsurface hydration products within the dentinal tubules	Control	15.5 (9.7 ˗ 20.3)	11.9 (7.2 ˗ 18.7)	11.1 (4.2 ˗ 27.9)	7.6 (3.6 ˗ 18.8)
	Early	15.5 (9.7 ˗ 20.3)	11.9 (5.7 ˗ 19)	11.1 (4.2 ˗ 27.9)	7.6 (3.6 ˗ 18.8)
	Delayed	13 (8.4 ˗ 32)	13 (7.5 ˗ 18.8)	9.3 (5.7 ˗ 36)	5.2 (5 ˗ 9.8)

Data are shown as median (minimum to maximum) values. Superscript lowercase letters in each column indicate significant differences among control, early, delayed groups according to adhesive application timing of CSBCs (P < 0.05)

The Ca at% did not significantly differ among subsurface hydration products within the dentinal tubules according to adhesive application timing for each CSBC material (P > 0.05). The Ca at% within the hydration products in dentinal tubules and POBS were not significantly related (P > 0.05; Spearman’s rank correlation analysis).

Discussion

Achieving durable adhesion between CSBCs and underlying dentin is essential to ensure marginal sealing and prevent restoration failure [23]. Before placing a definitive restoration, the applied biomaterial should achieve sufficient POBS to withstand the condensation pressure exerted during the restorative procedures [28]. The range of packing stress applied while placing a composite resin is 0.43 – 2.6 MPa [29]. Our POBS values of the materials were higher than these (Table 2), suggesting that they can withstand condensation forces.

Since the experimental set-up was intended for direct pulp capping, coronal dentin instead of root dentin was used in the present study. The mean POBS of the premixed-type CSBCs ranged 14.92–17.45 MPa, higher than the POBS of ProRoot MTA to root dentin with a thickness of 1 mm, reported to be 7.28–12.70 MPa [27, 30]. The tubule density, elastic modulus of tested material, and plunger pin diameter of the push-out test might affect the POBS [31, 32]. An increased number of open dentinal tubules can affect POBS [31]. Elastic modulus, mineral density and occlusion of dentinal tubules are influenced by aging [33]. Therefore, four tested materials were filled within each dentin disc to minimize the inter-tooth variability in dentin substrate properties.

The POBS values of all tested materials were lower in the early group, than in the delayed group, although only EC significantly differed. The null hypothesis that the timing of dentin adhesive application would not affect the push-out bond strength of CSBCs to dentin was rejected for EC but not for the remaining materials. The predominance of cohesive failure in the EC/early group might be due to the delayed maturation of the EC. In contrast, the TC group showed a higher incidence of cohesive failures irrespective of adhesive timing, suggesting a stable bond between the TC and dentin [34]. Nonetheless, failure mode analysis alone does not necessarily correlate with actual bond strength.

A universal dentin adhesive was applied one hour after CSBC placement in the early group. The initial setting times for EC, OF, and RM according to the manufacturers are below 10 min. However, the results of our pilot study revealed that when adhesive was applied 15–30 min after placement, the premixed-type CSBC surfaces were not sufficiently set and were dislodged upon air-drying to evaporate the solvent of the adhesive. This prevented us from measuring POBS. The expulsion of air-dried CSBC was rare after one hour. However, under clinical conditions, waiting for one hour during VPT is likely unfeasible. Therefore, delayed restoration, particularly when using EC, might be beneficial to achieve optimal POBS.

The null hypothesis that the type of calcium silicate-based cement would not affect the push-out bond strength of CSBCs to dentin was rejected. The POBS of TC was higher than those of the other tested materials, regardless of application timing. This might be attributed to its composition containing methacrylate monomers, which allow for rapid polymerization and improved physical strength [35, 36]. The results suggest that TC can be directly applied in cavity, allowing for permanent restoration in a single visit, thereby eliminating the need for temporary restoration. These results are in accordance with those of Candan et al. [24] and Shin et al. [37], who reported TC exhibited significantly higher SBS values than calcium silicate-based materials without resin content, such as NeoPutty and Well-Root PT.

The results of the SEM examination revealed spherical or cubic crystals on the surfaces of all tested CSBC materials at 7 days after placement without adhesive. This finding aligns with that of Chen et al. [38], who found uniformly shaped and larger particles in calcium silicate materials immersed in distilled water for 7 days. These transformed into plate-like crystalline structures and a denser layer as maturation progressed. Larger particles formed when the adhesive was applied at 1 h after filling with TC compared to the control. In contrast, crystals were denser in the TC of the delayed group when adhesive was applied 3 days after filling the CSBC, followed by maturation for 7 days. Consequently, this protocol provided the delayed specimens with extended maturation. While TC sets by light-curing, premixed CSBCs set by hydration. When applied 1 h after filling, the adhesive covered the surface of premixed CSBCs before they were completely set, which contributed to lower Ca at% in the EDS analysis. Phase separation or separated filler particles of the adhesive on the surface of premixed CSBCs were identified (Figs. 6B and 7B, and 8B). In contrast, crystalline maturation proceeded when the adhesive was applied 3 days after filling the premixed CSBCs (Figs. 6I and 7I, and 8I). These findings suggest that delaying the restoration may allow more favorable conditions for hydration, facilitating the formation of a denser and more homogeneous microstructure.

CSBCs promote calcium–phosphate precipitation at the dentin interface, making them favorable candidates for VPT applications [39]. This bioactive process leads to the formation of an interfacial layer with tag-like structures between MTA and dentin, thereby significantly enhancing the resistance to dislodgement [39–41]. Intratubular mineralization was denser in control and delayed than in the early groups (Fig. 9). The formation of crystalline structures might be hindered by adhesive penetration in the early groups. Furthermore, Namazikhah et al. [42] found that acidic environments might hinder crystal maturation, resulting in porous or needle-like fragmented structures. Although the less crystalline structures in the early groups were speculated to cause lower POBS, a quantitative analysis of intratubular mineralization could not be performed.

The present study evaluated restoration timing only up to three days, and POBS tests were conducted after seven days of storage. Longer waiting periods might further improve bonding between CSBCs and dentin and this should be investigated in future studies. Stress distribution generated during POBS tests might not be purely shear and can be affected by factors such as specimen geometry and plunger dimensions. Therefore, the measured values might not represent true interfacial shear strength. Nevertheless, because all specimens were tested under identical conditions, the model remained appropriate for relative comparisons among groups.

If completing the final restoration in a single visit is unavoidable, it may be considered to apply the adhesive primarily to the exposed tooth structure while minimizing its application on the CSBC surface, as a potential strategy to reduce CSBC exposure to the acidic environment. However, the effect of direct contact between composite resin and CSBC in the absence of adhesive was not evaluated in this study and warrants further investigation.

Limitations

This study has several limitations. We evaluated POBS after 7 days, thus representing only the early interfacial behavior. Long-term maturation periods might influence the extent of hydration and bond development. Furthermore, bond strength should be evaluated after the aging process to assess long-term performance and durability. The small pin size relative to the diameter of the filler CSBC might have caused material penetration rather than true interfacial debonding, which potentially affected the accuracy of the bond strength values [32].

The resolution of ×30 magnification in the failure mode analysis was insufficient to clearly distinguish thin adhesive remnants from cohesive failure. Accordingly, failure mode findings should be considered descriptive rather than definitive. Because one type of universal adhesive in self-etch mode was tested, the results should not be generalizable to other bonding systems or strategies [43, 44]. The present findings from this in vitro cannot be directly extrapolated to clinical practice.

Conclusions

Within the limitations of this in vitro study, adhesive application at 1 h after placement of calcium silicate-based cement adversely affected the 7-day push-out bond strength of a specific premixed calcium silicate-based cement. Delayed restoration might improve bond strength, particularly for Endocem MTA Premixed Heavy. Although POBS values were lower in premixed calcium silicate-based cements than in TheraCal PT, they maintained values exceeding 10 MPa. This suggests that premixed calcium silicate-based cements can withstand restorative procedures after pulp capping.

Abbreviations

10-MDP

10-methacryloyloxydecyl dihydrogen phosphate

at%

Atomic percentage

Bis-GMA

Bisphenol-A-glycidyl dimethacrylate

Ca

Calcium

CSBC

Calcium silicate-based cement

EC

Endocem MTA Premixed Heavy

MTA

Mineral trioxide aggregate

OF

One-Fil Putty

PENTA

Dipentaerythritol penta-acrylate phosphate

POBS

Push-out bond strength

RM

BC Universal RRM

TC

TheraCal PT

VPT

Vital pulp therapy

Authors’ contributions

Conception and design of the study: J.S. and S.O.; Acquisition of data: J.S., G.Y. and H.J.K.; Analysis of data: J.H.J., G.Y., S.H.C., S.W.C. and S.O.; Drafting manuscript: J.S. and G.Y.; Revising manuscript: H.J.K., J.H.J., S.H.C., S.W.C. and S.O. All authors reviewed the manuscript. J.S. and G.Y. contributed equally to this study as first authors.

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2025-00519698), and by the Bio & Medical Technology Development Program of the National Research Foundation (NRF) & funded by the Korean government (MSIP&MOHW) (No. RS2022NR071907).

Data availability

The corresponding author can provide the datasets used and/or analyzed during the current study upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the principles of the Declaration of Helsinki. All experimental protocols were approved by the Institutional Review Board of Kyung Hee University Dental Hospital. The requirement for informed consent to participate was waived by the Institutional Review Board of Kyung Hee University Dental Hospital (KH-DT24019), as the study utilized extracted teeth discarded as medical waste.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.