Introduction
The objective of this study was to investigate the in-vitro fluoride release from a glass ionomer orthodontic bonding system (Fuji I, GC Corporation, Tokyo, Japan) over a 2-month period and the in-vivo enamel fluoride uptake after 6 months.
Methods
Ten metal brackets were bonded with either glass ionomer or composite resin (Transbond XT, 3M Unitek, Monrovia, Calif; Light Cure), which served as controls, to recently extracted molars. The bonded teeth, cut at the level of their roots, were stored in distilled water that was renewed after every fluoride measurement at 1, 3, 7, 30, and 60 days. The in-vitro fluoride release was measured by using a fluoride ion-selective electrode, connected to an ion analyzer. Fifteen pairs of premolars were bonded with metal brackets with either the Fuji or the Transbond adhesive. Six months later, the teeth were extracted for orthodontic purposes, embedded in resin, and cross-sectioned, and the fluoride compositions between the outer and bulk enamel surfaces were evaluated with scanning electron microscopy and energy dispersive analysis. The results were analyzed with nonparametric 1-way analysis of variance (ANOVA) on ranks for in-vitro fluoride release and nonparametric 2-way ANOVA on ranks for in-vivo fluoride enamel uptake; group differences were investigated with the Holm-Sidak test at the .05 level. The Spearman rank correlation coefficient test was used to investigate the association between fluoride and aluminum levels in the interfaces of the specimens bonded.
Results
The initial burst of fluoride release observed for the Fuji adhesive after the first day of the experiment had a significant decrease with time, and it persisted throughout the monitoring period (60 days) ( P <0.05). Fluoride concentrations were found in both the outer and deeper enamel surfaces, with the outer sites having 4 times higher fluoride relative to the bulk for the glass ionomer ( P <0.05), and higher fluoride was found in the outer layers for the glass ionomer bonded enamel specimens ( P <0.05). However, the concurrent identification of aluminum and fluoride traces in the enamel implied that the source of this high fluoride concentration originated from cement particles and not from ionic uptake.
Conclusions
The short-term fluoride release and the absence of documented enamel uptake suggest that the glass ionomer orthodontic adhesive tested might provide protective action only through the reservoir mechanism.
Early enamel decalcification near orthodontic fixed appliances is a concern for orthodontists, especially in patients with severe crowding and poor oral hygiene. Clinically visible incipient caries lesions can develop within a month under poorly fitting molar bands or around brackets. Preventive measures, such as oral hygiene instructions and application of fluoride agents, have proven useful for protection against caries susceptibility. However, the effectiveness of these measures greatly depends on patients’ cooperation, and thus a bonding system capable of releasing fluoride on a long-term basis would be extremely beneficial.
The anticariogenic potential of glass ionomer cements, with their ability to release fluoride, has been shown in several in-vitro studies. However, there is limited in-vivo evidence regarding the amount of fluoride released from this material and subsequently its protective role, and this is mainly derived from short-term monitoring periods. Therefore, there is a need for long-term in-vivo studies to determine the mechanism of protective action of the fluoride released by the bonding material.
The aims of this study were to investigate over a period of 2 months the fluoride released from a glass ionomer cement in vitro and to assess the fluoride uptake from the enamel bonded to brackets with a glass ionomer in vivo after 6 monthsof service intraorally. The testing hypotheses were that fluoride is released in vitro from glass ionomer cement during a 2-month period, and that fluoride is taken up by the enamel of the teeth bonded with the same adhesive material after 6 months in vivo.
Material and methods
Ten metal brackets (Mini Sprint, Forestadent, Pforzheim, Germany) in 2 groups of 5 were bonded with either glass ionomer cement (Fuji I, GC Corporation, Tokyo, Japan) or composite resin (Transbond XT, 3M Unitek, Monrovia, Calif) to the middle buccal enamel surface of premolars extracted for orthodontic purposes. The roots of the extracted premolars were cut under continuous water flow before their use. After bonding, the brackets were left intact to be stored in 10 mL of fresh distilled water, which was renewed after every fluoride measurement in plastic vials at 37°C.
All fluoride analyses were carried out by using a fluoride ion specific electrode (type 9609 BN, Orion Research, Beverly, Mass) connected to an expandable ion analyzer (type EA940, Orion Research). Fluoride measurements (ppm of fluoride) were taken at days 1, 3, 7, 30, and 60. After 6 consecutive measurements, the electrode was calibrated with a series of standard fluoride solutions (0.01, 0.1, 1, 10, and 100 ppm of fluoride), and fresh distilled water was measured as the control. For fluoride measurements, 1 mL of sample solution was transferred into a 5-mL polyethylene test cup. An equal amount (1 mL) of total ionic strength adjustment buffer aluminum was added to each sample to stabilize the pH and allowed to adjust to room temperature of 25°C (temperature of the fluoride electrode’s optimal function). After each fluoride measurement of the sample solution, the volume of that initial solution was decreased by 1 mL.
The sample for this study consisted of patients from the Graduate Orthodontic Clinic of Aristotle University of Thessaloniki, Greece, with an age range of 13 to 36 years (mean, 22 years) and a male-to-female ratio of 0.55. They were selected from a larger pool of participants by using the following selection criteria: need for extractions for orthodontic purposes of either 2 or 4 premolars, intact premolar buccal surfaces, and no previous glass ionomer restorative fillings in the oral cavity. The study was approved by the university’s ethical committee, and, after detailed explanations of the purpose of this research and the clinical procedures, consent was obtained.
During the 6-month monitoring period, the patients were asked to refrain from any additional fluoride supplements except for their regular fluoride toothpaste for oral hygiene. Furthermore, no orthodontic treatment would begin until after the extraction of the premolars, 6 months after their bonding with the adhesives to be tested.
The teeth were randomly classified in 15 contralateral pairs, cleaned, and polished with a nonfluoride paste (Clean Polish, Hawe-Neos Dental, Bioggio, Switzerland). Metallic brackets (Mini Sprint, Forestadent) were bonded with either glass ionomer cement (Fuji I) or composite resin (Transbond XT) following the manufacturers’ instructions, with the split-mouth design. Before resin bonding, the enamel was etched with 35% phosphoric acid gel (Transbond XT etching gel) for 30 seconds. After 6 months, the teeth were carefully extracted, with the brackets intact on their buccal surface, and thoroughly rinsed with distilled water. An hour after storage in wet cotton with distilled water, the extracted premolars were embedded in epoxy resin (CaldoFix, Struers A/S, Ballerup, Denmark) in plastic cylindrical boxes, with the proximal surfaces parallel to the horizontal plane. The specimens were sectioned buccolingually with a microtome by using an Isomet low speed saw (11-1180, Buehler, Lake Bluff, Ill) and sputter-coated with carbon (SCD 004 sputter coater with CEA 035 Carbon Evaporation Supply, Bal-Tec, Balzers, Liechtenstein).
The surfaces of the cross-sections were studied under the scanning electron microscope (Quanta 200, FEI, Hillsboro, Ore), and the elemental composition was determined by energy-dispersive x-ray microanalysis with a silicon (lithium) detector (Sapphire, EDAX, Mahwah, NJ) with a super ultra-thin beryllium window. For each specimen, 2 spectra were collected in the outer enamel surface (2 μm below the outer enamel surface) and 100 μm near the enamel-dentin junction. The spectra were collected with 10 kV accelerating voltage, 110 μA beam current, and 1800-second acquisition time operating in spot analysis mode. The quantitative analysis of the percentage of weight concentration was performed by Genesis software (version 5.1, EDAX) with a nonstandard analysis mode by using ZAF correction methods.
Statistical analysis
For the in-vitro part of the study, the data were analyzed by using nonparametric 1-way analysis of variance (ANOVA) on ranks, with time as a predictor; further intergroup differences were investigated with the Holm-Sidak test. For the in-vivo part of the study, a nonparametric 2-way ANOVA on ranks was used, with enamel site (outer vs deeper layer) and material as the discriminants. Further intergroup differences were investigated with the Holm-Sidak test. The Spearman rank correlation coefficient test was used to investigate the association of fluoride and aluminum levels in the interfaces. All statistical analyses were carried out at the 95% level of significance with the NC SS 2007 (Statistical & Power Analysis Software, Kaysville, Utah) statistical package.
Results
In the in-vitro study, the rate of fluoride release (ppm) from glass ionomer cement with time is shown in Figure 1 . Table I presents the means and standard deviations of the fluoride released from the orthodontic adhesives and the statistical significance of the differences. The initial high fluoride release of the first day of the experiment decreased sharply to almost half in 3 days, and then continued to drop to a third in 7 days, a seventh in 30 days, and a fourteeth in 60 days. As expected, composite resin released no traces of fluoride and was only used as the control (not shown).
| Time | Fluoride release (ppm) Mean (SD) | Holm-Sidak grouping |
|---|---|---|
| Day 1 | 12.00 (0.01) | A |
| Day 3 | 5.23 (0.21) | B |
| Day 7 | 3.70 (0.20) | C |
| Day 30 | 1.70 (0.01) | D |
| Day 60 | 0.84 (0.02) | E |
In the in-vivo study, representative secondary electron images of the cross-sections of metal brackets bonded with glass ionomer or composite resin are shown in Figure 2 . Typical x-ray energy dispersive analysis spectra regarding the fluoride concentration and recorded at the outer (interface) and deeper (bulk) enamel surfaces are shown in Figure 3 . From these spectra, it is clear that, in the enamel specimens bonded with the glass ionomer, the concentration of fluoride in the outer enamel surface is much higher than in the deeper surface. In the control specimens bonded with composite resin, there was a variation in the fluoride concentration between the 2 enamel levels, but this difference did not reach statistical significance.
In the 15 specimens bonded with glass ionomer cement, fluoride was detected both at the outer 2 μm of enamel surface and at 100 μm from the dentin-enamel junction, with the outer enamel site having 4 times higher fluoride concentration than the deeper ( Table II ) ( P <0.05). A significant difference in fluoride concentration was confirmed between the outer and deeper enamel surfaces ( P <0.05) for the glass ionomer specimens, but not for the composite resin. When testing the outer 2 μm of enamel surface of the adhesives in the study, a statistically significant difference was found between the glass ionomer and the composite resin controls ( P <0.05), whereas no difference was demonstrated in the bulk of the enamel (at 100 μm from the dentin-enamel junction) between the 2 adhesives. The Spearman rank correlation coefficient test in the glass ionomer specimens showed positive correlation (r s = 0.791, P = 0.000) between the fluoride and aluminum concentrations ( Fig 4 ), implying the presence of cement particles.
| Enamel site | Enamel bonded with GI (wt %) | Enamel bonded with CR (wt %) |
|---|---|---|
| Outer enamel surface (2 μm) | 0.33 (0.22) a1 | 0.14 (0.09) b1 |
| Deeper enamel surface (100 μm from dentin-enamel junction) | 0.08 (0.05) a2 | 0.06 (0.02) a1 |
Results
In the in-vitro study, the rate of fluoride release (ppm) from glass ionomer cement with time is shown in Figure 1 . Table I presents the means and standard deviations of the fluoride released from the orthodontic adhesives and the statistical significance of the differences. The initial high fluoride release of the first day of the experiment decreased sharply to almost half in 3 days, and then continued to drop to a third in 7 days, a seventh in 30 days, and a fourteeth in 60 days. As expected, composite resin released no traces of fluoride and was only used as the control (not shown).
