Fluoride-Releasing Restorative Materials: The Effect of a Resinous Coat on Ion Release.

OBJECTIVE: To determine the effect of two adhesive systems and a glass ionomer coating resin on fluoride release and concurrent pH changes over a period of 168 days.
MATERIAL AND METHODS: Four restorative materials were investigated: a giomer Beautiful II, an "alkasite" material Cention, a conventional composite Filtek Z250, and a glass ionomer cement Fuji IX Extra. Light-cured composite specimens were coated using G-aenial Bond and Clearfil Universal Bond Quick. Glass ionomer specimens were coated using GC Fuji Coat LC. Uncoated specimens were used as references. Quantitative fluoride release and pH changes were measured after1 h, 24 h, 2 days, 7 days, 28 days, 84 days, and 168 days.
RESULTS: The cumulative fluoride release after 168 days increased for uncoated specimens in the following order: Filtek Z250 < Beautifil II < Cention < Fuji IX Extra. A comparatively lower fluoride release was measured for the composites coated with Clearfil Universal Bond Quick, with cumulative values after 168 days increasing in the following order: Filtek Z250 < Beautifil II < Cention. The composites coated with G-aenial Bond showed lower fluoride release compared to the uncoated specimens, with cumulative values increasing in the following order: Filtek Z250 < Beautifil II < Cention. The composites coated with G-aenial Bond showed pH values in the acidic range (4.4-5.7) after 1 h and 24 h.
CONCLUSION: Fluoride release varied among the investigated restorative materials and depended on the use of dental adhesives and coatings. The pH of all materials, coating types and time points varied.

Introduction

Glass ionomer cements (GICs), giomers, and composites are capable of releasing fluoride ions; therefore, they are used in restorative dental medicine for temporary and permanent fillings (). Fluoride ions have been incorporated into dental materials. They have been shown to produce a remineralizing effect and reduce cariogenic potential by acting on the growth and metabolism of S. mutans (-).

The release of fluoride ions from dental materials depends on factors related to an individual material and factors related to oral environment. The characteristics of materials such as composition, filler content, powder/liquid ratio, mixing process, and surface exposed to the aqueous medium affect the release of fluoride ions. Environmental factors include the pH and composition of the immersion medium and application of a dental adhesive system or coat (, -).

Previous studies showed a negative effect of a dental adhesive system or coat on the quantity of released fluoride ions. This has been attributed to the adhesive system or coat forming a hydrophobic barrier and thereby diminishing the diffusion of fluoride ions. This effect was more pronounced in resin composites than in GICs, which depended on type of the adhesive system or coat (, -).

Apart from the well-known classes of restorative materials that are capable of releasing fluoride ions (GICs and giomers), a novel “alkasite” restorative material has recently been introduced to the dental market. This bioactive material is capable of releasing remineralizing calcium and fluoride ions and can also neutralize acid by releasing OH- ions. It is compositionally similar to the group of composite materials and is designed to be used with or without an adhesive system, depending on the operator’s preferences (, ).

The effect of contemporary universal adhesives on fluoride release from remineralizing resin composite materials has not been investigated up to date. Therefore, the aim of this study was to compare fluoride release from specimens of restorative materials that were either coated or uncoated with a resinous layer of an adhesive system (for composite materials) and coat (for a GIC). Additionally, the effects of a resinous layer on the pH changes of the immersion medium were investigated. The first null hypothesis was that there would be no difference in fluoride ion release and pH changes between coated and uncoated specimens. The second null hypothesis was that the release of fluoride ions and pH changes of the immersion medium did not differ among the investigated restorative materials.

Material and methods

Restorative materials, adhesive systems, and a glass ionomer coat investigated in this study are presented in Table 1. Six specimens were prepared for each experimental group (n = 6).

Table 1

Materials used in this study

Material class	Commercial name (abbreviation)	Composition	Manufacturer	Shade /LOT No.
Giomer	Beautifil II (BF)	S-PRG (surface pre-reacted glass ionomer)Nano fillers 83.3 wt%	Shofu Dental GmbH, Ratingen, Germany	A2/051829
Alkasite	Cention (CN)	Filler: calcium fluorosilicate glass, Ba-Al silicate glass, Ca-Ba-Al fluorosilicate glass, Ytterbium trifluoride, IsofillerMonomers: urethane dimethacrylate UDMA, Tricyclodecan-dimethanol dimethacrylate DCP, Aromatic-aliphatic-UDMA, Polyethylene glycol 400 dimethacrylate PEG-400 DMAInitiator system: hydroperoxide, Ivocerin and acyl phosphine oxide	Ivoclar Vivadent, Schaan, Liechtenstein	A2/XL7102
Glass ionomer	Fuji IX Extra (FUJ)	Liquid: 5-10% polybasic carboxylic acidPowder: glass, oxide	GC Europe, Leuven, Belgium	A3/1801171
Conventional composite	Filtek Z250 (FIL)	Matrix: Bisphenol A diglycidyl ether dimethacrylate (BIS-GMA) and triethylene glycol dimethacrylate (TEGDMA), UDMAFiller: zirconia and silica particles 78.5 wt %, 60% vol.	3M Deutschland GmbH, Neuss, Germany	A2/N984652
Universal adhesive	G-aenial Bond (GB)	acetone 25-50%dimethacrylate 10-20%phosphoric acid ester monomer 5-10%dimethacrylate component 1-5%photoinitiator 1-5%butylated hydroxytoluene (BHT) < 0.5%	GC Europe, Leuven, Belgium	1811281
Universal fluoride-releasing adhesive	Clearfil Universal Bond Quick (CB)	BIS-GMA 10-25%, ethanol 10-25%, 2-hydroxyethyl methacrylate 2.5-10%, 10-Methacryloyloxydecyl dihydrogen phosphate, hydrophilic amide monomers, colloidal silica, silane coupling agent, sodium fluoride, dl-Camphorquinone, water	Kuraray Europe, Hattersheim am Main, Germany	3L0108
Glass ionomer coat	GC Fuji Coat LC (FC)	methyl methacrylate (MMA) 25-50%photoinitiator 1-5%butylated hydroxytoluene (BHT) < 1%	GC Europe, Leuven, Belgium	1804021

Specimens of the composite materials were prepared using cylindrical Teflon molds with a diameter of 6 mm and a height of 2 mm. The molds were placed on a polyethylene terephthalate (PET) foil, filled with uncured material and covered with another layer of PET foil (). The excess material was removed and specimens were polymerized using an LED curing unit (Bluephase G2, Ivoclar Vivadent, Schaan, Liechtenstein) with a nominal intensity of 1200 mW/cm2 for 20 s on each side. Specimens of the GIC were cast into the mold described above, covered with PET films, and left to set for 6 min according to manufacturer’s instructions (). To ensure that the specimens were completely surrounded by an aqueous medium, a plastic thread incorporated within each specimen was used to hang the specimens from the cap of the vial.

The schematic overview of the study design is presented in Figure 1. Each sample was immersed separately in a plastic vial containing 5 ml of deionized water at 37 °C and evaluated after 0, 1, 2, 7, 28, 84, and 168 days. At each time interval, the specimens were removed from the immersion medium, placed in 5 ml of new deionized water, and stored in the incubator at 37 °C until the next time point. The pH of the immersion medium was measured using the pH meter MP 220 (Mettler Toledo, Columbus, Ohio) and the InLab Expert Pro pH electrode (Mettler Toledo, Columbus, Ohio). Prior to the pH measurement, the electrode was calibrated using standard buffer solutions at​​ pH = 4 and pH = 7.

Figure 1

Schematic representation of the study design.

A volume of 4.5 ml of the medium and 0.5 ml of TISAB III buffer (Total Ionic Strength Adjustment Buffer, Thermo Fisher Scientific, Chelmsford, USA) were placed in a new beaker to determine the fluoride ions concentration. Quantitative fluoride ions release was measured using a standard ion-selective electrode Orion 9609BNWP (Thermo Fisher Scientific, Massachusetts, USA), ISO 19448: 2018. Before measurements, the ion-selective electrode was calibrated using a series of standards of known concentration in a range of 10−5–10−2 mol/L F−.

The fluid samples were placed on a RH magnetic stirrer (IKA-Werke GmbH & Co. KG, Staufen, Germany), which was set at 500 rpm without heating. A sensor was placed inside the fluid sample, which rotated at the indicated speed during the measurement. An ion-selective electrode was connected to an Expandable Ion Analyzer EA 940 (Orion Research, Beverly, USA) from which the values ​​were read. An average of 3 readings was calculated for each sample.

Normality of distribution was verified using the Shapiro Wilk’s test, and equality of variances was checked using the Levene’s test. Mean values of fluoride release and pH changes of the immersion medium were compared among the combinations of materials and resinous coatings using a one-way ANOVA with Tukey post-hoc adjustment. Mean values of fluoride release and pH changes of the immersion medium for each combination of materials and resinous coatings were compared among different measurement times using repeated-measures ANOVA with Bonferroni post-hoc adjustment. For all analyses, the overall level of significance was set at 0.05. Statistical analysis was performed in SPSS 20 (IBM, Armonk, NY, USA).

Results

Fluoride release

Mean concentrations of released fluoride ions from uncoated specimens are presented in Figure 2. All uncoated groups showed a statistically significant increase in released fluoride ions over time in the following order: BF < CN < FUJ. When comparing cumulative values after 168 days, FUJ demonstrated 3 times higher values than CN, and 35 times higher values than BF.

Figure 2

Fluoride ions release from uncoated specimens. Uppercase letters denote statistically homogeneous groups within materials. Lowercase letters denote statistically homogeneous groups within time points. Error bars denote one standard deviation.

The effect of coating systems on fluoride release is shown in Table 2. The coating of composite specimens with adhesive systems led to a statistically significant decrease in the amount of released fluoride ions. Materials treated with GB released more fluoride ions than materials treated with CB. FIL released fluoride ions only when fluoride-releasing adhesive CB was applied. Both FUJ groups showed a statistically significant increase in released fluoride ions over time, whereas uncoated specimens reached 30 times higher values compared to the specimens coated with FC.

Table 2

Fluoride ions release from coated and uncoated specimens. Uppercase letters denote statistically homogeneous groups within materials subgroups (with or without coat). Lowercase letters denote statistically homogeneous materials subgroups (time points). Values are presented in ppm. Standard deviations are presented in parentheses.

Time (days)		FILTEK Z250			BEAUTIFIL II			CENTION			FUJI IX EXTRA
	uncoated		Clearfil Universal Bond Quick	G-aenial Bond	uncoated	Clearfil Universal Bond Quick	G-aenial Bond	uncoated	Clearfil Universal Bond Quick	G-aenial Bond	uncoated	GC Fuji COAT LC
0	0		0 A	0	0.05 (0.03) Aa	0.026 (0.02) Aab	0.01 (0.01) Ab	0.08 (0.02) Aa	0.004 (0.001) Ab	0.005 (0.002) Ab	2.95 (0.5) A	0.18 (0.06) A
1	0		0 A	0	0.11 (0.05) Aa	0.026 (0.02) Ab	0.013 (0.01) Ab	4.22 (0.44) Ba	0.057 (0.02) Ab	0.386 (0.2) Ab	7.95 (0.89) B	0.26 (0.1) A
2	0		0 A	0	0.14 (0.06) Aa	0.026 (0.02) Ab	0.013 (0.05) Ab	7.15 (0.64) Ca	0.071 (0.02) ABb	0.462 (0.22) Ab	9.94 (0.84) C	0.18 (0.12) A
7	0		0.0000833 A	0	0.22 (0.07) Aa	0.025 (0.02) Ab	0.016 (0.04) Ab	8.93 (0.75) Da	0.08 (0.03) ABCb	0.91 (0.33) Ac	14.5 (0.56) D	0.61 (0.23) AB
28	0		0.000695 B	0	0.6 (0.16) Ba	0.043 (0.01) Ab	0.087 (0.01) Ab	10.12 (0.99) DEa	0.23 (0.05) BCb	2.669 (0.72) Bc	22.49 (1.16) E	0.97 (0.33) BC
84	0		0.0007667 B	0	1.01 (0.26) Ca	0.08 (0.03) Bb	0.322 (0.07) Bc	10.8 (1.06) Ea	0.24 (0.05) Cb	3.088 (0.69) Bc	45.7 (2.12) F	1.18 (0.42) C
168	0		0.0007667 B	0	2.19 (0.4) Da	0.08 (0.03) Bb	1.77 (0.17) Bc	12.75 (1.07) Fa	0.80 (0.23) Db	5.072 (0.99) Cc	55.49 (3) G	1.85 (0.5) D

pH changes

pH changes of uncoated specimens are presented in Figure 3. The pH of dental materials differed in the first measurement (1 h time point) in the following order: FIL < FUJ < CN < BF (6.16, 6.47, 6.83, and 7.26). Also, the values ​​differed at the last measurement (168 days): FUJ < FIL < CN < BF (6.56, 6.63, 6.91, and 7.45).

Figure 3

pH changes in uncoated specimens. Lowercase letters denote statistically homogeneous groups within materials. Uppercase letters denote statistically homogeneous groups within time points. Error bars denote one standard deviation.

pH values in coated and uncoated specimens are presented in Table 3. pH values showed growth tendency over time in all tested materials. FIL and CN showed higher pH values when coated with CB than the uncoated group. BF showed the highest values in the uncoated group. FUJ showed higher values when coated with FC. GB showed lower pH values in all groups after 1 hour and 24 hours. After 1 hour, the pH values among the materials increased in the following order: FIL < CN < BF (4.42, 4.53, and 4.75). After 24 hours, the pH values increased among the materials in the following order: FIL < BF < CN (5.39, 5.67, and 5.73).

Table 3

pH changes in coated and uncoated groups. Uppercase letters denote statistically homogeneous groups within materials subgroups (with or without coat). Lowercase letters denote statistically homogeneous materials subgroups within time points. Values are presented in ppm. Standard deviations are presented in parentheses.

Time (days)		FILTEK Z250			BEAUTIFIL II			CENTION			FUJI IX EXTRA
	uncoated		Clearfil Universal Bond Quick	G-aenial Bond	uncoated	Clearfil Universal Bond Quick	G-aenial Bond	uncoated	Clearfil Universal Bond Quick	G-aenial Bond	uncoated	GC Fuji COAT LC
0	6.16 (0.08) Aa		6.41 (0.2) Ab	4.42 (0.08) Ac	7.27 (0.16) Aa	6.64 (0.33) Ab	4.75 (0.11) Ac	6.83 (0.11) ABCa	7 (0.2) ABCa	4.53 (0.16) Ab	6.47 (0.27) A	6.42 (0.08) A
1	6.26 (0.04) Aa		7.63 (0.32) Cb	5.39 (0.74) Bc	6.53 (0.06) Ba	6.61 (0.14) Aa	5.67 (0.21) Bb	6.99 (0.2) Ca	6.52 (0.19) Ab	5.73 (0.38) Bc	6.81 (0.33) AB	6.89 (0.19) B
2	7.5 (0.32) Da		7.5 (0.4) Ca	6.89 (0.23) Cb	6.57 (0.24) Ba	6.58 (0.09) Aa	6.9 (0.15) CDb	6.61 (0.15) Aa	6.92 (0.51) Aba	6.79 (0.25) Ca	7.11 (0.2) B	6.96 (0.22) B
7	6.96 (0.21) Ca		6.8 (0.15) ABab	6.64 (0.23) Cb	6.99 (0.2) Aa	6.9 (0.13) ABa	7.44 (0.27) Db	6.87 (0.26) ABCa	7.55 (0.25) CDb	7.19 (0.22) Cab	6.76 (0.1) AB	7.64 (0.29) C
28	6.96 (0.09) Ca		6.49 (0.11) Ab	6.4 (0.09) Cb	7.49 (0.18) Ca	7.03 (0.12) BCb	7.48 (0.35) Da	6.68 (0.19) ABa	7.23 (0.33) BCDa	6.85 (0.58) Ca	6.57 (0.16) A	7.79 (0.22) C
84	6.86 (0.12) BCab		6.97 (0.1) Bb	6.69 (0.15) Ca	7.48 (0.31) Ca	7.19 (0.26) BCa	7.19 (0.41) CDa	6.78 (0.1) ABCa	7.62 (0.26) Db	6.91 (0.36) Ca	6.63 (0.12) A	7.87 (0.23) C
168	6.63 (0.04) Ba		6.97 (0.09) Bb	6.74 (0.12) Ca	7.47 (0.17) Ca	7.27 (0.24) Ca	6.71 (0.55) Cb	6.91 (0.08) BCa	7.4 (0.37) BCDb	6.94 (0.12) Ca	6.56 (0.25) A	7.89 (0.15) C

Discussion

This study investigated fluoride release of four restorative materials and the effect of contemporary universal adhesive systems, and a coat on fluoride ions release. Also, for all coated and uncoated specimens, pH changes were evaluated. According to a statistically significant effect of the material type and resinous coating on fluoride release and pH changes, both null hypotheses were rejected.

All fluoride-releasing materials used in this study showed long-term fluoride release, as reported in several previous studies (, ). FUJ presented the highest cumulative values of released fluoride ions, followed by CN and BF. As expected, the conventional composite FIL did not release fluoride ions, except when its specimens were coated with the fluoride-releasing adhesive CB.

GICs are characterized by the initial release of a large amount of fluoride ions called the burst effect which occurs within 24 h of cement setting time (-). According to Wiegand et al., this effect results from a reaction between glass particles and polyalkenoate acid ().

Our study confirmed a similar burst effect in FUJ specimens, and also in the alkasite composite material CN. However, the burs effect was present only in uncoated FUJ and CN specimens. The observed burst effect found in CN could be a result of its composition. According to the respective manufacturer, this material comprises 78.4 wt% of the following inorganic fillers: barium aluminum silicate glass, ytterbium trifluoride, Isofiller (patented filler), calcium barium aluminum fluorosilicate glass, and a calcium fluorosilicate (alkaline) glass. Also, 24.6 wt% of the material is composed of alkaline (calcium fluorosilicate) glass filler, which is responsible for fluoride, hydroxide, and calcium ions release ().

Our results are partially in contrast with those from a study of Gupta et al., which tested similar materials, Cention N (self- and light-cured) and a conventional GIC. Their results showed a time-dependent decrease in fluoride release from all tested materials, except for GIC in an acidic immersion medium. This difference between the released fluoride ions could be a result of different measuring time periods in the aforementioned study (7, 21, and 28 days). However, their results of a GIC in neutral medium releasing significantly higher amounts of fluoride ions than Cention N corroborate our results of fluoride ions release in these materials ().

Tiskaya et al. evaluated fluoride release, pH changes, and apatite formation of two bioactive composites (Cention N and Activa). Their cumulative values of fluoride ions release were below 8 ppm after 42 days, which is lower than the values obtained in our study (30.49 ppm after 28 days). This difference could be due to different specimen geometry and different immersion media. In our study, specimens were stored in deionized water while artificial saliva was used in the aforementioned study. Specimens immersed in artificial saliva tend to show a 17–25% lower fluoride release compared to specimens immersed in water (, ). This is explained by a lower diffusion gradient between the material and artificial saliva compared to the diffusion gradient between the material and deionized water. Also, artificial saliva may contain components that form a pellicle on the material surface and thereby interfere with ion release, decreasing it for about 15-20% (, - ).

The giomer BF did not show the burst effect, which is in correspondence with the study of Yap et al. (). The slower release of fluoride ions from BF can be attributed to its hydrophobic resinous matrix and a relatively low amount of pre-reacted glass ionomer fillers. Similar to the findings of our study, Mousavinasab et al. have also reported a higher release of fluoride ions from a GIC compared to a giomer. According to their study, the differences in fluoride ions release between those materials could be caused by a greater porosity of GIC, lack of glass ionomer matrix phase, and incorporated resin components in giomers ().

In our study, the giomer showed the lowest cumulative fluoride release. According to a study by Colceriu Burtea et al., this finding can be explained by material characteristics where giomers include polyacrylic acid from PRG (pre-reacted glass) rather than amino acid modified polyalkenoic acid in the composition of the PRG. They also found that experimental giomers which contained hydrophilic and flexible polymer matrix based on UDMA showed higher cumulative fluoride ions release than giomers based on rigid and hydrophobic dimethacrylates (TEGDMA in BF) (, , , ).

The results of our study showed that the amount of released fluoride ions in all investigated materials was diminished in coated specimens, which is in line with previous studies (, , ). Surprisingly, the fluoride-releasing adhesive system reduced the amount of leached fluoride ions.

In the present study, the materials coated with GB showed higher fluoride ions release than materials coated with the fluoride-releasing adhesive system CB. This could be explained by material composition. Water sorption depends on material hydrophilicity and can reduce polymer mechanical properties affecting its hygrothermal degradation and polymer hydrolysis that later forms water channels, surface erosions, and crazing that impact material permeability (-). High concentrations of acidic monomers contribute to hydrophilicity. GB contains 5-10% phosphoric acid ester monomers, while CB contains hydrophilic amide monomers (, ). Resin monomers with ester bonds are highly prone to hydrolysis in the presence of water, which could be one of the reasons for an increased fluoride ions release from materials coated with GB (, ). Also, some studies found that resin polarity can act as a major determinant of water uptake. An increase in polarity results in higher water sorption. Polar functional groups include OH- groups, carboxyl groups and phosphate groups, which tend to form hydrogen bonds with water. The water molecules that are „bound” by polar functional groups induce swelling and plasticization of the polymer network (, -). The GB adhesive system used in this study had dimethacrylate (10–20%) and dimethacrylate components (1-5%). Those components could partially explain an increased amount of released fluoride ions from materials coated with GB.

The hydrophilic resins used in universal adhesive systems are prone to limited monomer conversion due to phase separation which leads to degradation in an aqueous medium (, ). Oguri et al. showed that degree of conversion could depend on the functional monomer and photoinitiator system (). Also, the lack of compatibility between hydrophobic photoinitiator and hydrophilic monomers showed lower values of degree of conversion when compared with hydrophilic photoinitiator and monomers. A lower degree of conversion can affect adhesive permeability and lead to acidic monomer diffusion (-). CB scientific documentation declares the presence of hydrophobic photoinitiator camphorquinone and hydrophilic amide monomers which could explain lower pH values than pH values in uncoated specimens after 1 hour for BF and FIL, and after 24 hours for CN (). A lower degree of conversion can also be speculated to have led to lower pH values for GB at time intervals 1 hour and 24 hours.

According to its respective manufacturer, CN shows a buffering ability by releasing acid-neutralizing hydroxide ions. Our results demonstrated small pH changes when CN was exposed to neutral medium, which can be compared with the studies of Gupta et al. and Tiskaya et al. (, ). Also, an interesting finding from our study was that the average pH level for CN was higher in the group that was coated with CB compared with the uncoated group. A similar result was present for FIL, but not for BF, and this could be due to increased reactivity when the acidic coat was applied.

FUJ also presented higher pH values when coated with FC. Given that uncoated FUJ specimens tended to decrease the pH value from neutral to acidic over time, the barrier formed in coated specimens showed a lack of that tendency.

GB showed lower pH levels than uncoated specimens and coated with CB. This can be related with a higher concentration of released fluoride ions when compared with materials coated with CB adhesive. The results of similar studies showed that the highest fluoride ions release was found in acidic medium (, -).

Conclusions

The amount of released fluoride ions varied among dental materials and depended on the use of adhesive systems and coatings. The glass ionomer Fuji IX Extra showed the highest values of released fluoride ions followed by the alkasite material Cention and the giomer Beautifil II. Both adhesive systems and the coat had a diminishing effect on released fluoride ions. pH values of the immersion medium differed among materials, treatments and time points. The amount of released fluoride ions showed a growth tendency over time in all tested materials. The lowest pH values were identified in all material specimens coated with G-aenial Bond.