How effectively do hydraulic calcium-silicate cements re-mineralize demineralized dentin

Highlights

  • Hydraulic calcium-silicate cements (hCSCs) do re-mineralize demineralized dentin.

  • The resin-free hCSCs induced re-mineralization at a higher speed/intensity than the resin-based cement.

  • Re-mineralization was incomplete for all hCSCs tested, this even at 6 months.

Abstract

Objective

To characterize the chemical interplay and to quantify the re-mineralization potential of hydraulic calcium-silicate cements (hCSCs) at demineralized dentin.

Methods

Pairs of class-I cavities were prepared in non-carious human third molars. One dentin cavity was demineralized with 10% formic acid (5 h); the other served as control. The cavities were filled with two resin-free hCSCs (Biodentine, Septodont; ProRoot MTA, Dentsply Sirona) or one resin-based hCSC (TheraCal LC, Bisco). After 1-week, 1-, 3-, and 6-month storage in simulated body fluid (SBF), polished specimen cross-sections were chemically characterized using Field-emission-gun Electron Probe Micro-Analysis (Feg-EPMA) and micro-Raman spectroscopy (μRaman).

Results

Feg-EPMA line-scans and elemental mappings confirmed early re-mineralization induced by all three hCSCs at 1 week. The relative depth and intensity of re-mineralization were for the resin-free hCSCs in the range of 50.5%–84.8% and 68.1%–89.2%, respectively. Re-mineralization did not significantly differ for the storage periods (p > 0.05). Significantly less re-mineralization was achieved by the resin-based hCSC TheraCal LC that reached only at 6 months a re-mineralization level that was no longer significantly different from that achieved by the resin-free hCSCs at 1 week (p > 0.05). Re-mineralization of intertubular dentin, including tubular occlusion, was observed; Si was occasionally detected to have infiltrated the dentin tubules. Dentin re-mineralization by hCSCs was confirmed using μRaman that revealed an increased phosphate peak at 960 cm −1 .

Significance

hCSCs do re-mineralize demineralized dentin. The resin-free cements induced re-mineralization at a higher speed/intensity than the resin-based hCSC. However, re-mineralization was incomplete for all hCSCs tested, this even at 6 months.

Introduction

Hydraulic calcium-silicate cements (hCSCs), being widely used for various pulp-related and endodontic indications , are currently also materials of interest to serve as dentin re-mineralization agents. Re-mineralization of artificial caries-like demineralized dentin has been achieved in a study using a so-called ‘biomimetically modified mineral trioxide aggregate’ (MTA mixed with biomimetic analogs such as polyacrylic acid and sodium tripolyphosphate) in the presence of simulated body fluid (SBF) . In addition to the primary pulp-related/endodontic indications, the commercially available hCSC Biodentine (Septodont, Saint Maur des Fosses, France) has also been claimed for direct restorative applications, such as permanent dentin-replacement . Its capability to re-mineralize demineralized dentin has been explored . Another study evaluated the effectiveness of Biodentine as an indirect pulp-capping material; it revealed a success rate of 83.3% in patients with reversible pulpitis after 12-month follow-up . Besides resin-free hCSCs, light-curing resin-based calcium-silicate cements, such as TheraCal LC (Bisco, Schaumburg, IL, USA), have been developed for indirect (and direct) pulp capping as well.

To date, the dentin re-mineralization potential of hCSCs reported in literature can only be considered as ‘proof-of-concept’, so far without direct clinical implementation potential. Laboratory studies conducted by various research groups reporting on the dentin re-mineralization potential of hCSCs utilized thinly sectioned dentin slabs rather than that they subjected dentin exposed in actually prepared tooth cavities to re-mineralization protocols. In some studies, the dentin slabs were immersed in biomimetic analogs containing SBF to generate fluid nano-precursors . Hence, there remains a need to characterize the re-mineralization potential of hCSCs under more clinically simulated conditions. Using a tooth-cavity model, demineralized dentin should be exposed to the cements, like we characterized in a previous study the interfacial interaction of hCSCs with unaffected sound dentin .

The purpose of this study was therefore to characterize the chemical interplay and to quantify the re-mineralization effectiveness of two resin-free hCSCs (Biodentine, Septodont; ProRoot MTA, Dentsply Sirona) and one resin-based hCSC (TheraCal LC, Bisco), when applied onto demineralized dentin in tooth cavities; the specimens were stored in SBF for different time periods. The null hypotheses tested were (1) that there is no difference in the chemical interplay of the resin-free and resin-based hCSCs with demineralized dentin, (2) that all hCSCs effectively (completely) re-mineralize demineralized dentin, and (3) that the re-mineralization potential does not depend on the SBF-storage period.

Materials & methods

Specimen preparation

Thirty-six healthy human third molars (gathered as approved by the Commission for Medical Ethics of KU Leuven under the file number S57622) were stored in 0.5% chloramine solution at 4 °C and were used within 3 m after extraction. Each tooth was mounted in a gypsum block to facilitate manipulation. The occlusal third of the crown was removed using a diamond saw (Isomet 1000, Buehler, Lake Bluff, IL, USA). The dentin surfaces were verified for absence of enamel and/or pulp tissue under a stereo-microscope (Wild M5A, Heerbrugg, Switzerland). The surfaces were then coated with a layer of bonding agent from the self-etch adhesive Clearfil SE Bond (Kuraray Noritake, Tokyo, Japan). A standard box-type class-I cavity with the floor ending at mid-coronal dentin (3 × 1.5 mm wide, 0.5 mm deep) was prepared in each tooth using a medium-grit (107 μm) diamond bur (842, Komet, Lemgo, Germany) fixed in a water-cooled high-speed turbine mounted in a custom-adapted MicroSpecimen Former (University of Iowa, Iowa City, IA, USA). Cavity surfaces were verified for absence of pulp tissue under the stereo-microscope (Wild M5A), upon which the cavity was filled with 10% formic acid (pH 2) to demineralize the cavity surfaces; this demineralization process lasted for 5 h. The tooth was kept humid by immersing the bottom of the gypsum block in water. After that, another cavity (3 × 1.5 mm wide, 0.5 mm deep) was prepared on the same tooth next to the previous one at the area that was beforehand protected against demineralization by the bonding agent. This cavity, in which unaffected sound dentin was exposed, served as control. Both cavities were next rinsed using a strong spray with distilled water for 10 s and air-dried, after which they were filled with one of the two resin-free hCSCs Biodentine (Septodont) or ProRoot MTA (Dentsply Sirona), or with the resin-based hCSC TheraCal LC (Bisco) ( Table 1 ). A plastic spatula (provided with Biodentine, Septodont) was used to condense the cement, while holding the gypsum blocks with the embedded teeth on a vibrating table (Porex, Aachen, Germany) to ensure proper adaptation of the hCSC to the dentin cavity walls. The resin-based hCSC TheraCal LC (Bisco) was light-cured following the manufacturer’s instructions for 20 s (Bluephase 20i, Ivoclar Vivadent, Schaan, Liechtenstein; light output of 1200 mW/cm 2 , as measured with a Marc Resin Calibrator of BlueLight Analytics, Halifax, Canada). The teeth restored with Biodentine (Septodont) and ProRoot MTA (Dentsply Sirona) were stored in a sealed container and were kept humid by immersing the bottom of the gypsum blocks in water, respectively, for 12 min and 4 h, this according to the initial setting time of the cements, as mentioned by the respective manufacturers.

Table 1
List of the cements investigated.
Cement Manufacturer Lot number Composition a
Biodentine Septodont B07023 Powder: tricalcium silicate, dicalcium silicate, calcium carbonate and oxide, iron oxide, zirconium oxide
Liquid: distilled water, calcium chloride, hydrosoluble polymer
ProRoot MTA Dentsply Sirona 0000116362 Powder: tricalcium silicate, dicalcium silicate, tricalcium aluminate, bismuth oxide
Liquid: distilled water
TheraCal LC b Bisco 1400003407 Portland cement type III, polyethylene glycol dimethacrylate, barium zirconate

a The composition is based on technical information provided by the respective manufacturer.

b A light-curing resin-modified calcium–silicate cement.

Upon setting, all teeth were removed from the gypsum blocks and immediately immersed in SBF at 37 °C for 1 w, 1, 3 and 6 m (n = 3 per experimental group and storage period). SBF consisted of 136.8 mM NaCl, 4.2 mM NaHCO 3 , 3.0 mM KCl, 1.0 mM K 2 HPO 4 ·3H 2 0, 1.5 mM MgCl 2 ·6H 2 O, 40 mM HCl, 2.5 mM CaCl 2 , 0.5 mM Na 2 SO 4 , 50 mM (CH 2 OH) 3 CNH 2 ; its pH was adjusted to 7.4. After storage, the teeth were cross-sectioned perpendicular to the cement-dentin interface by means of a water-cooled diamond saw (Isomet 1000, Buehler). All specimens were subsequently processed for electron microscopy by fixation in 2.5% glutaraldehyde for 24 h, dehydration in ascending concentrations of ethanol (25, 50, 75, and 95% for 30 min each, and lastly 100% for 1 h), and finally drying by immersion in hexamethyldisilazane (HMDS) for 10 min, this following a method described in detail by Perdigão et al. . Upon drying, both the ‘cement-unaffected sound dentin’ and the ‘cement-demineralized dentin’ interfaces were polished using an argon-ion-beam (IB-09010CP Cross Section Polisher, Jeol, Tokyo, Japan) at 5.0 kV for 7 h to achieve an ion-beam polished interfacial area of approximately 1 mm 2 .

Field-emission-gun Electron Probe Micro-Analysis (Feg-EPMA)

The ion-beam polished interfaces were coated by a 2-nm thick platinum-palladium (Pt-Pd) layer using a turbomolecular-pumped coater (Q150T S, Quorum, East Sussex, UK). In each specimen, the intensities of chemical elements (Ca, P) along the interface were quantified along three 180-μm long line-scans using a Field-emission-gun Electron Probe Micro-Analyzer (Feg-EPMA; JXA-8530F, Jeol, Tokyo, Japan) at a spatial resolution of ±0.05 μm. The results of the Feg-EPMA line-scans analyses were imported into a software package (R3.01, R Foundation for Statistical Computing, Vienna, Austria). The position, from where dentin was demineralized and re-mineralized, was indicated by lines I and line II, respectively. The position, until where dentin was re-mineralized and originally demineralized was indicated by the lines III and IV, respectively. The depth of demineralization was determined as the distance between line I and IV in μm using an automated script; the re-mineralization depth, if existed, was determined as the distance between line II and III, this when the Ca and P concentrations were increased as compared to those within demineralized dentin. The relative re-mineralization depth (D RM ) was calculated as the ratio of the re-mineralization depth to the initial depth of demineralized dentin. To evaluate to which level re-mineralization was achieved, the relative remineralization intensity (I RM ) was calculated as the percentage of the mean P intensity in re-mineralized dentin as compared with that in deeper unaffected sound dentin.

In case chemical changes were detected at the cement-dentin interface, representative points were selected and the elemental composition (Ca, P, C, Si) of these points were quantitatively analyzed, based on which also the Ca/P weight ratio was calculated. In addition, the interface of one representative specimen per experimental group (cement) and storage period was chemically mapped for Ca, P, C and Si.

X-ray profiles and element quantifications were performed at 15 kV (voltage) and 15 μA (probe current) under high vacuum. No peak overlapping was detected.

Micro-Raman spectroscopy (μRaman)

In case any chemical interfacial changes were detected by Feg-EPMA, the corresponding interface was polished again using the argon-ion-beam polisher (IB-09010CP Cross Section Polisher, Jeol) at 5.0 kV for 7 h to remove the Pt–Pd coating. Five structurally different points were randomly selected within the chemical interaction zone and were subsequently analyzed using μRaman (Senterra, Bruker, Billerica, MA, USA). μRaman assessment was performed using a near-infrared (785 nm) laser, a 50× microscope objective and 50 × 1000 aperture, this at an integration time of 10 s with 3 co-additions. For Biodentine (Septodont) and TheraCal LC (Bisco) specimens, the laser power was set to 100 mV, while the power was reduced to 50 mV for ProRoot MTA (Dentsply Sirona) specimens, as this cement was sensitive to laser damage (observed in our previous study ). The CCD detector possessed a 1024 × 256 pixel resolution and was cooled down thermo-electrically to a temperature of −65 °C. Post-processing of the spectra was conducted using the Opus Spectroscopy Software version 7.0 (BrukerOptik, Ettlingen, Germany); concave rubberband baseline correction was conducted at an iteration number of 8.

Statistical analysis

D RM and I RM were statistically analyzed by Kruskal Nemenyi multiple comparison tests to assess the effects of cement type (two resin-free cements: Biodentine, Septodont; ProRoot MTA, Dentsply Sirona; one resin-based cement: TheraCal LC, Bisco) as well as storage period. Tests were performed at a significance level of α = 0.05 using a software package (R3.01).

Materials & methods

Specimen preparation

Thirty-six healthy human third molars (gathered as approved by the Commission for Medical Ethics of KU Leuven under the file number S57622) were stored in 0.5% chloramine solution at 4 °C and were used within 3 m after extraction. Each tooth was mounted in a gypsum block to facilitate manipulation. The occlusal third of the crown was removed using a diamond saw (Isomet 1000, Buehler, Lake Bluff, IL, USA). The dentin surfaces were verified for absence of enamel and/or pulp tissue under a stereo-microscope (Wild M5A, Heerbrugg, Switzerland). The surfaces were then coated with a layer of bonding agent from the self-etch adhesive Clearfil SE Bond (Kuraray Noritake, Tokyo, Japan). A standard box-type class-I cavity with the floor ending at mid-coronal dentin (3 × 1.5 mm wide, 0.5 mm deep) was prepared in each tooth using a medium-grit (107 μm) diamond bur (842, Komet, Lemgo, Germany) fixed in a water-cooled high-speed turbine mounted in a custom-adapted MicroSpecimen Former (University of Iowa, Iowa City, IA, USA). Cavity surfaces were verified for absence of pulp tissue under the stereo-microscope (Wild M5A), upon which the cavity was filled with 10% formic acid (pH 2) to demineralize the cavity surfaces; this demineralization process lasted for 5 h. The tooth was kept humid by immersing the bottom of the gypsum block in water. After that, another cavity (3 × 1.5 mm wide, 0.5 mm deep) was prepared on the same tooth next to the previous one at the area that was beforehand protected against demineralization by the bonding agent. This cavity, in which unaffected sound dentin was exposed, served as control. Both cavities were next rinsed using a strong spray with distilled water for 10 s and air-dried, after which they were filled with one of the two resin-free hCSCs Biodentine (Septodont) or ProRoot MTA (Dentsply Sirona), or with the resin-based hCSC TheraCal LC (Bisco) ( Table 1 ). A plastic spatula (provided with Biodentine, Septodont) was used to condense the cement, while holding the gypsum blocks with the embedded teeth on a vibrating table (Porex, Aachen, Germany) to ensure proper adaptation of the hCSC to the dentin cavity walls. The resin-based hCSC TheraCal LC (Bisco) was light-cured following the manufacturer’s instructions for 20 s (Bluephase 20i, Ivoclar Vivadent, Schaan, Liechtenstein; light output of 1200 mW/cm 2 , as measured with a Marc Resin Calibrator of BlueLight Analytics, Halifax, Canada). The teeth restored with Biodentine (Septodont) and ProRoot MTA (Dentsply Sirona) were stored in a sealed container and were kept humid by immersing the bottom of the gypsum blocks in water, respectively, for 12 min and 4 h, this according to the initial setting time of the cements, as mentioned by the respective manufacturers.

Table 1
List of the cements investigated.
Cement Manufacturer Lot number Composition a
Biodentine Septodont B07023 Powder: tricalcium silicate, dicalcium silicate, calcium carbonate and oxide, iron oxide, zirconium oxide
Liquid: distilled water, calcium chloride, hydrosoluble polymer
ProRoot MTA Dentsply Sirona 0000116362 Powder: tricalcium silicate, dicalcium silicate, tricalcium aluminate, bismuth oxide
Liquid: distilled water
TheraCal LC b Bisco 1400003407 Portland cement type III, polyethylene glycol dimethacrylate, barium zirconate

a The composition is based on technical information provided by the respective manufacturer.

b A light-curing resin-modified calcium–silicate cement.

Upon setting, all teeth were removed from the gypsum blocks and immediately immersed in SBF at 37 °C for 1 w, 1, 3 and 6 m (n = 3 per experimental group and storage period). SBF consisted of 136.8 mM NaCl, 4.2 mM NaHCO 3 , 3.0 mM KCl, 1.0 mM K 2 HPO 4 ·3H 2 0, 1.5 mM MgCl 2 ·6H 2 O, 40 mM HCl, 2.5 mM CaCl 2 , 0.5 mM Na 2 SO 4 , 50 mM (CH 2 OH) 3 CNH 2 ; its pH was adjusted to 7.4. After storage, the teeth were cross-sectioned perpendicular to the cement-dentin interface by means of a water-cooled diamond saw (Isomet 1000, Buehler). All specimens were subsequently processed for electron microscopy by fixation in 2.5% glutaraldehyde for 24 h, dehydration in ascending concentrations of ethanol (25, 50, 75, and 95% for 30 min each, and lastly 100% for 1 h), and finally drying by immersion in hexamethyldisilazane (HMDS) for 10 min, this following a method described in detail by Perdigão et al. . Upon drying, both the ‘cement-unaffected sound dentin’ and the ‘cement-demineralized dentin’ interfaces were polished using an argon-ion-beam (IB-09010CP Cross Section Polisher, Jeol, Tokyo, Japan) at 5.0 kV for 7 h to achieve an ion-beam polished interfacial area of approximately 1 mm 2 .

Field-emission-gun Electron Probe Micro-Analysis (Feg-EPMA)

The ion-beam polished interfaces were coated by a 2-nm thick platinum-palladium (Pt-Pd) layer using a turbomolecular-pumped coater (Q150T S, Quorum, East Sussex, UK). In each specimen, the intensities of chemical elements (Ca, P) along the interface were quantified along three 180-μm long line-scans using a Field-emission-gun Electron Probe Micro-Analyzer (Feg-EPMA; JXA-8530F, Jeol, Tokyo, Japan) at a spatial resolution of ±0.05 μm. The results of the Feg-EPMA line-scans analyses were imported into a software package (R3.01, R Foundation for Statistical Computing, Vienna, Austria). The position, from where dentin was demineralized and re-mineralized, was indicated by lines I and line II, respectively. The position, until where dentin was re-mineralized and originally demineralized was indicated by the lines III and IV, respectively. The depth of demineralization was determined as the distance between line I and IV in μm using an automated script; the re-mineralization depth, if existed, was determined as the distance between line II and III, this when the Ca and P concentrations were increased as compared to those within demineralized dentin. The relative re-mineralization depth (D RM ) was calculated as the ratio of the re-mineralization depth to the initial depth of demineralized dentin. To evaluate to which level re-mineralization was achieved, the relative remineralization intensity (I RM ) was calculated as the percentage of the mean P intensity in re-mineralized dentin as compared with that in deeper unaffected sound dentin.

In case chemical changes were detected at the cement-dentin interface, representative points were selected and the elemental composition (Ca, P, C, Si) of these points were quantitatively analyzed, based on which also the Ca/P weight ratio was calculated. In addition, the interface of one representative specimen per experimental group (cement) and storage period was chemically mapped for Ca, P, C and Si.

X-ray profiles and element quantifications were performed at 15 kV (voltage) and 15 μA (probe current) under high vacuum. No peak overlapping was detected.

Micro-Raman spectroscopy (μRaman)

In case any chemical interfacial changes were detected by Feg-EPMA, the corresponding interface was polished again using the argon-ion-beam polisher (IB-09010CP Cross Section Polisher, Jeol) at 5.0 kV for 7 h to remove the Pt–Pd coating. Five structurally different points were randomly selected within the chemical interaction zone and were subsequently analyzed using μRaman (Senterra, Bruker, Billerica, MA, USA). μRaman assessment was performed using a near-infrared (785 nm) laser, a 50× microscope objective and 50 × 1000 aperture, this at an integration time of 10 s with 3 co-additions. For Biodentine (Septodont) and TheraCal LC (Bisco) specimens, the laser power was set to 100 mV, while the power was reduced to 50 mV for ProRoot MTA (Dentsply Sirona) specimens, as this cement was sensitive to laser damage (observed in our previous study ). The CCD detector possessed a 1024 × 256 pixel resolution and was cooled down thermo-electrically to a temperature of −65 °C. Post-processing of the spectra was conducted using the Opus Spectroscopy Software version 7.0 (BrukerOptik, Ettlingen, Germany); concave rubberband baseline correction was conducted at an iteration number of 8.

Statistical analysis

D RM and I RM were statistically analyzed by Kruskal Nemenyi multiple comparison tests to assess the effects of cement type (two resin-free cements: Biodentine, Septodont; ProRoot MTA, Dentsply Sirona; one resin-based cement: TheraCal LC, Bisco) as well as storage period. Tests were performed at a significance level of α = 0.05 using a software package (R3.01).

Results

The elemental composition of unaffected sound and demineralized dentin

Representative EPMA line-scan analysis along the demineralized dentin together with the corresponding BSE image, is shown in Fig. 1 . Along this demineralized dentin, both Ca and P contents were lowest at the most superficial demineralized dentin, which had been in the closest contact with 10% formic acid; the Ca and P contents increased gradually towards the inner deeper demineralized dentin zone ( Fig. 1 b).

Fig. 1
Feg-EPMA Line scan chemical analysis across demineralized dentin upon 5 h exposure to 10% formic acid. The analysis was performed along the red/white line superimposed on the respective Feg-SEM photomicrograph (a). The Ca and P contents increased gradually from the superficially demineralized dentin, which had been in direct contact with 10% formic acid, towards the inner deeper demineralized dentin zone, thereby confirming that dentin was effectively demineralized (b). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Representative Feg-EPMA mappings of Ca, P and C at unaffected sound and demineralized dentin, as well as the corresponding back-scattered electron (BSE) photomicrographs are shown in Fig. 2 . The elemental composition of the points indicated in the BSE images ( Fig. 2 a and e) was quantified and mentioned in Table 2 .

Nov 22, 2017 | Posted by in Dental Materials | Comments Off on How effectively do hydraulic calcium-silicate cements re-mineralize demineralized dentin
Premium Wordpress Themes by UFO Themes