Fluoride release from conventional, resin-modified and hybrid glass ionomer cements

Summary Introduction The aim of the study was to quantify and compare fluoride release from four different glass ionomer cement products (GICs). Materials and Methods Standardized disk-shaped samples (5x2mm; n=5/group) of GIC restorative materials: conventional (Fuji IX, GC Corp., Japan), resin-modified (Fuji II LC, GC) and hybrid glass ionomer cement (Equia Forte, GC) and a conventional GIC liner/base material (Alfagal, Galenika, Serbia) were tested for fluoride release up to 21 days postsetting. Each sample was immersed in 5 mL of fresh deionized water during each time interval. Universal microhybrid composite (Filtek Z250, 3M EPSE, USA) and adhesive (Adper Single Bond, 3M ESPE) were used as negative controls. Fluoride release was measured using an F-selective electrode (Cole-Parmer, USA) and an ion meter (Oakton 700, Cole-Parmer, USA). Data were statistically analyzed using one-way ANOVA, regression and correlation analysis at 0.05. Results The highest total fluoride release was measured from Alfagal (386±61 ppm/g), and significantly less from Equia (188±29 ppm/g), Fuji IX (143±11 ppm/g) and Fuji II LC (104±14 ppm/g) (p < 0.05). All GICs showed the highest fluoride release during the first 24 hours post-setting. After 3 days, fluoride release slowed down reaching a plateau for all materials. Regression and Pearson correlation analysis showed significant inverse relationship between fluoride release and sample mass and density (p<0.001). Conclusion Of the three GICs indicated for use as restorative materials, Equia Forte released the highest fluoride concentration. Fluoride release was material and density dependent, with higher release occurring from lower density GICs.


INTRODUCTION
One of the main reasons for restoration failure is secondary caries [1,2,3] which results in further tooth tissue loss, weakened remaining tooth structure or even premature tooth loss. Any mechanism that inhibits acid production by bacteria, increases resistance to demineralization or facilitates remineralization could be considered clinically significant. A recent study showed that the most commonly used restorative materials in contemporary restorative dentistry, resin-based composites, do not exhibit buffering potential, thus being more susceptible to the formation of secondary caries around such restorations [4].
Fluoride from fluoride-containing materials has the capacity to chemically interact with hydroxyapatite of enamel and dentin adjacent to the restoration resulting in the formation of fluorapatite [5,6] which increases resistance to acid demineralization and prevents secondary caries [7,8]. Fluoride released from fluoride-containing restorative materials, such as glass ionomer cements (GICs), may improve resistance to demineralization [9,10], facilitate remineralization [11,12] or even directly affect cariogenic bacteria by inhibiting their metabolic enzymes [13]. Fluoride was found in dentin, released from the bottom parts of GIC restorations (Fuji IX and Fuji II LC improved) in artificially demineralized monkey`s teeth after three days [14]. Fluoride-releasing materials show cariostatic properties and may affect bacterial metabolism under simulated cariogenic conditions in vitro [15].
GICs have been substantially improved over the years, especially their mechanical properties, thereby expanding indications for use. These improvements include the introduction of photopolimerizable resins (light-cured GIC), or ultrafine, highly reactive glass particles, dispersed within the conventional glass ionomer structure and a higher molecular weight polyacrylic acid in hybrid GIC. The latter has led to the recently launched hybrid GIC, Equia Forte, currently the only material in this 'class' of GICs.
It has been widely accepted that during the setting action of GICs a variety of ionic constituents is released from the glass phase, including fluoride. Fluoride is released in short-term by rapid dissolution from the outer surface of the set material into solution. Sustained ion release is the consequence of ion diffusion through the bulk cement. Despite the latest improvements, it is imperative that GICs maintain abundant fluoride release in order to support their antiocariogenic potential.
The aim of the present study was to quantify and compare fluoride release from the conventional, resin-modified and the new glass hybrid GICs into deionized water 1, 6, 24 h, 3, 7, 14 and 21 days post-setting. The null hypotheses were: (1) there is no significant difference in fluoride release from different GICs and (2) there is no significant relationship between fluoride release and sample mass/density.

MATERIALS AND METHODS
Details on the materials used in this study are given in the Table 1. All materials were used according to manufacturers' instructions. Capsules of Fuji IX, Fuji II and Equia were mixed in an auto-mixer for 10 s. Alfagal was prepared by hand mixing 1 scoop of powder and 2 drops of liquid for 30 s using a plastic spatula on the paper pad. Z250 and Adper were used directly from the tube or bottle, respectively.
Standardized plastic molds, 5 mm in diameter and 2 mm deep, were placed on a Mylar strip and a glass pad, filled with GIC, composite or adhesive, covered with another strip, and pressed with a glass slide to extrude excess material and form smooth surfaces. Five samples were prepared per group, except the adhesive control group (Adper) with one prepared sample to avoid wasting material. Alfagal and Equia were allowed to set for 6 min and 2 min 30 s, respectively. Fuji II, Z250 and Adper were light-cured through the Mylar strip and 1 mm thick glass slide using a conventional LED light-curing unit (LEDition, Ivoclar Vivadent, Schaan, Liechtenstein) operating at intensity of 800 mW/cm 2 . Adper and Z250 were lightcured for 40 s, and Fuji II for 20 s each from the top and bottom side. Each sample was weighed on an analytical balance with an accuracy of 0.1 mg (ACCULAB ALC-110.4, Sartorius group, Goettingen, Germany).
All samples were stored dry for 24 h at 37°C. Following storage, each sample was immersed in 5 mL of deionised water in a sterile glass vial and kept at 37°C. Fluoride concentrations were measured after 1 h, 6 h, 24 h, 3, 7, 14 and 21 days using a F-selective electrode (Cole-Parmer, Bunker CT, Vernon Hills, Illinois) and an ion meter (Oakton pH/Ion 700 Bench Meter, Cole-Parmer, Bunker CT, Vernon Hills, Illinois). The electrode was first calibrated using 0.1, 1 and 10 ppm F. Ionic Strength Adjuster solution (0.5 mL) was added to all tested solutions just before measuring. Between measurements the electrode was rinsed with deionized water.
Statistical analysis was performed in Minitab 16 (Minitab Inc., State College, PA, USA). The data were analyzed using one-way ANOVA with Tukey's post-hoc test for multiple comparisons at the level of significance alpha=0.05. Regression analysis and Pearson correlation were performed to determine the relationship between fluoride release and sample mass/density.

RESULTS
Aside from the control composite Z250, the mass and density of GIC samples varied significantly, with the conventional GIC Alfagal showing lower mass and density than other tested GICs (p<0.001) ( Table 2). Fluoride release varied in concentration between materials and time periods ( Figure 1). All GICs had the highest fluoride release during the first 24 h post-setting. Relatively high fluoride release occurred from GICs except Fuji II over the next 48 h (the first 3 days in total) as indicated by steep slopes in Figure 1. Further release was notable until the end of the experiment (21 days) with base/liner GIC Alfagal showing the highest fluoride release in all tested times (more than double the values of other materials) (p<0.05). Equia released more fluoride than Fuji IX and Fuji II with Fuji II releasing the least fluoride concentration of the tested GICs (p<0.05). Fluoride release in negative control groups, both composite Z250 and Adper adhesive, was 10 or more times lower compared to GICs (11.46±2.45 and 14.81 ppm per gram composite and adhesive, respectively, over 21 days).
Regression and Pearson correlation analysis showed significant inverse relationship between fluoride release (F) and sample mass and density (p<0.001) ( Figure 2). The regression equations may be expressed as Equations 1 and 2: with R-sq = 86.52% and R-sq(adj.) = 85.77% Pearson correlation coefficient r=-0.930 and p<0.001 indicate strong negative correlation between fluoride release and mass/density i.e. higher fluoride release occurred from lower density GICs.

DISCUSSION
Both null hypotheses were rejected as the results confirmed significant differences in fluoride release between GICs as well as an inverse relationship between fluoride release and sample mass and density.
In the present study, a range of GICs with different composition was tested, from the 'classical' GIC formulated Alfagal to reinforced conventional Fuji IX, resin-modified and light-curable Fuji II to the latest glass hybrid Equia. This choice was made to cover a wide range of different GIC compositions so as to ascertain a range of potential fluoride concentrations released over a period of 21 days.
There is a lack of standardization vis-à-vis sample size, shape, the type and quantity of immersion media as   well as data presentation in the current literature. Data comparison is often difficult due to these differences. The present results are expressed as the amount of fluoride released from 1 gram of material. Deionized water was used as the immersion medium as it is most often used in other studies and has been shown to facilitate more fluoride release than artificial saliva [16].
Short-and long-term fluoride release from restorative materials is related to their matrices, setting mechanisms and fluoride content and as well as environmental conditions [15]. The present results showed the highest fluoride release from the conventional GIC Alfagal. Alfagal has the most 'classical' GIC composition of all tested GICs, based on a water solution of acrylic and itaconic acid copolymers. It is of lower density (lower viscosity) than other tested GICs and is specifically indicated for use as a liner/ base under direct and indirect restorations. Alfagal is not indicated as a filling (restorative) material, not even in non-load-bearing areas, in contrast to other tested GICs. It is well known that early, conventional GICs exhibit inferior mechanical properties than other filling materials and profound sensitivity to water imbalance, especially during the first 24 h [17]. Resin modification of the conventional formula led to somewhat improved mechanical properties but still below those of resin-based composites [18]. Also, GICs with higher powder-to-liquid ratio exhibit better mechanical properties [19]. Higher water uptake and solubility as well as less complicated internal structure compared to GICs containing high molecular polyacrylic and polybasic carboxylic acid and/or resin monomers may have led to more pronounced fluoride release from Alfagal than other tested GICs.
Equia had higher fluoride release compared to Fuji IX and Fuji II, but the difference was not statistically significant between Equia and Fuji IX, probably due to relatively high SD values. However, the results are indicative of a tendency of higher fluoride release from Equia than Fuji IX. These two GICs share a similar composition, and Equia is considered a successor of Fuji IX. Higher fluoride content, slightly lower density or other compositional modification undisclosed by the manufacturer could be the reason(s) for somewhat higher fluoride release from Equia than Fuji IX. In clinical practice it is recommended to cover the surface of Equia restorations with Equia Forte Coat, a light-curable resin-based liquid. This coat would probably act as a semipermeable membrane, allowing partial fluoride release into the oral environment. However, low wear resistance of unfilled or very low filled resin liquid indicates that protective coat would be worn during function leaving Equia exposed for unrestricted fluoride release. As the longevity of the coat layer in clinical conditions is unpredictable and individual, the present study design without any protective layer allowed measuring maximum fluoride release for the given sample size and shape.
Previous studies reported different findings related to fluoride release from resin-modified and conventional GICs. Several studies showed no significant difference in fluoride release between light-cured and conventional GICs [20,21,22]. The present study detected less fluoride release from light-cured Fuji II compared to other tested GICs. Fluoride release from resin-modified GICs could be hampered by the polymer network intertwined with polyalkeenoate chains.
Following the highest fluoride release during the first 24h post-setting, fluoride release from each material decreased sharply over the first week and continued to decrease steadily over the 3 weeks period which is in agreement with other studies reporting the maximum release during the first 24-48 h [23][24][25][26][27]. In all GICs a tendency for fluoride release was observable based on the increasing slope in Figure 1. This indicated that new formulations of GICs also act as a pool of fluoride with potential continuous release over a long time, especially in environment with acidic pH. Earlier studies have shown a steady fluoride release over the period of 2 years from conventional GICs [28].
Continuous fluoride release was also detected in a rare in vivo study by Koch et al. [29] who showed that fluoride concentration in saliva immediately after placement of GIC restorations increased from 0.04 ppm to 0.8-1.2 ppm, but slowly decreased about 35 % after 3 weeks and additional 30 % after 6 weeks.
Although clinical importance of fluoride as an anticariogenic agent is nowadays generally accepted, the evidence to corroborate this statement comes from in vitro and in situ studies. Randomized clinical trials offer inconclusive evidence of greater caries protection by GICs [30]. Ex vivo studies showed the potential of fluoride ions to migrate from GIC restorations into the surrounding enamel and dentin of primary molars [31,32]. Also, GICs were shown to inhibit secondary caries formation in vitro in artificial biofilm models [33,34]. Until clinical evidence becomes definitive regarding the anticariogenic efficiency of fluoride-containing materials, primarily GICs, it is important that new and improved formulations of these materials maintain high levels of fluoride release.

CONCLUSION
Conventional, glass hybrid and resin-modified GICs showed continuous release of fluoride ions over 21 days after setting. Concentrations of released fluoride differed and were affected by material composition and density. The addition of resin into GIC formulation decreased its ability to release fluoride. Higher viscosity GICs could be also linked to lower fluoride release. Of the three GICs indicated for use as restorative materials, Fuji II LC, Fuji IX and Equia Forte, the highest fluoride release occurred from Equia Forte.

ACKNOWLEDGEMENT
This study was supported by Research grants ON172007 from the Ministry of Education, Science and Technological Development, Republic of Serbia.