INTRODUCTION
Consumption of aqai (Euterpe olerácea) is popular among Brazilian children, teenagers and adults, especially in the summer. The fruit is rich in compounds with variable colours 1 (green, red, and purple) and the pH of commercial aqai pulps is low 2 . These characteristics might be a source of potential adverse effects for aesthetic dental materials, particularly conventional glass ionomer cement 3 . The anticaries properties 4 and substantial evidence confirm the outstanding performance of resin modified glass ionomer cement (RMGI) for sealing5and restoring 6 primary teeth. However, some level of discolouration and increase in roughness are likely 7 , even though it is more resistant to physical and mechanical changes than its conventional counterpart. The hydrophilic monomers in RMGI composition, such as HEMA (hydroxyethyl methacrylate), BIS-GMA (bisphenol A-glycidyl methacrylate), and TEGDMA (triethylene glycol dimethacrylate), may be prone to discolouration and increased roughness, shortening the material’s longevity 8 , 9 .
Although paediatric patients care about aesthetic issues, it is not reasonable for dentists to spend time polishing or even replacing restorations because of aqai-related stains (particularly in low-compliance children). Given the lack of evidence on RMGI and aqai interaction, information on it is relevant. Thus, the aim of this study was to assess the effect of aqai on the colour and surface roughness of an RMGIC in vitro. The null hypothesis tested was that aqai would not affect these variables.
MATERIAL AND METHOD
Sample Size Calculation
The sample size of 66 specimens (22 per group) was calculated in terms of the difference among the three groups (Test F; one-way ANOVA) for colour change magnitude (ΔE*), a error level of 5%, effect size of 0.4, and P error level of 20% (software GPower 3.1.9.2, University of Düsseldorf). Given potential losses during specimen preparation, 10% (25 per group) was added.
Experimental Design
Seventy-five disc-shaped specimens (2 mm thick, 8 mm in diameter) of Vitremer, PEDO shade (3M ESPE, St Paul, MN, USA; Batch 1927700210) ( Table 1 ) 10 , 11 , 12 were randomly distributed into three groups (n=25) corresponding to the tested solutions: artificial saliva (control), aqaí sorbet and aqai juice. The chemical degradation protocol was based on Ozera et al. (2019) 13 . CIE L*a*b* and surface roughness (Ra, mm) values were analysed at baseline (L*, a*, b*, Ra) and after degradation (L’*, a’*, b’*, Ra’). The mean values of each specimen were recorded.
Material and Specimen Preparation
A trained operator hand mixed Vitremer's powder/ liquid (1:1 ratio) following the manufacturer’s instructions. Silicone moulds (2 mm deep, 8 mm internal 0) (Adsil Silicone Addition, Coltene, Vigodent, Rio de Janeiro, Brazil) were filled slowly with the material using a Centrix syringe (Dentsply Ind. E Com. Ltda., Petrópolis, RJ, Brazil). The mould was placed between two mylar strips, and the top was pressed by hand with a glass plate to make the specimens flat and smooth 14 . The glass was removed, and the mixture was light-cured with Valo (Ultradent, USA) under 1,000 mW/cm2 for 20 s, following the manufacturer’s instructions. After the material had set and the strips had been removed, a thin layer of glaze (the finishing gloss included in the Vitremer kit) was applied to protect the specimens, followed by light curing, as described above 15 . A radiometer (Hilux Dental Curing Light Meter, Benlioglu Dental Inc., Demetron, Ankara, Turkey) was used to monitor the irradiance before and after three measurements. The specimens were not subjected to finishing and polishing procedures. Finally, all 75 specimens were placed in deionized water at 37°C for 24 h to set (manufacturer’s recommendation) 14 .
Material | Vitremer |
---|---|
Manufacturer | 3M ESPE (St Paul, MN, USA) |
Batch # | 1927700210 |
Powder content | Fluoroaluminosilicate glass; redox system. Mean Filler Size 3.0 μm. |
Liquid content | Copolymer of Acrylic and Itaconic Acids, water, HEMA, Ethyl Acetate, Diphenyliodonium Hexafluorophosphate. |
Glaze content | Triethylene Glycol Dimethacrylate (TEGDMA), Bisphenol A Diglycidyl Ether Dimethacrylate (BISGMA), Triphenylantimony, 4-(Dimethylamino)-Benzeneethanol, Hydroquinone. |
Colour and roughness assessments
A CM-2600d/2500d spectrophotometer (Konica Minolta, Tokyo, Japan) was used to record the mean values of each colour coordinate (measured in triplícate). The ΔE* calculation followed the formula ΔE*= [ΔL*2 + Δa*2 + Δb*2] ‘/í13. Placing the specimens on a white background prevented potential absorption effects. Surface roughness was measured with a Surfcorder SE 1700 instrument (Kosaka Corp, Tokyo, Japan) to record the mean values (Ra, mm) from three successive measurements at the centre of each specimen in different directions, covering 1.25 mm with a cut-off length of 0.25 mm at a tracing speed of 0.1 mm/s 15 .
Solutions and Chemical Degradation Protocol
Table 2 provides the composition, manufacturer, batch number and pH of the solutions (artificial saliva, sorbet, and juice). The decrease in the pH of aqai observed after 7 days in the pilot study required the use of fresh solutions. The specimens were subjected to three daily soaks (6 ml, 15 minutes) for 14 days at 37°C. After these soaks and a wash with distilled water, the specimens were placed in fresh saliva for 30 minutes. After the third soak, the specimens were placed in fresh saliva, where they remained until the next day 13 . A SC06 electrode (Sensoglass, SensopH Ind. E Comércio de Sensores, Sao Paulo) coupled to an ion 450 M analyser (Analyser Analytical Instrumentaron, Sao Paulo, Brazil) was used to check the pH before beginning the test.
Statistical analysis
Statistical analysis was performed with Jamovi 1.2. software (5% significance). The Shapiro-Wilk and Levene tests were used to assess data normality and homogeneity. The variables L*, a*, b*, b’*, ΔL*, and ΔE* were subjected to ANOVA followed by Tukey’s test. The other variables (Ra, Ra’, ARa, L’*, a’*, Δa* and Δb*) were subjected to the Kruskal-Wallis test followed by the Dwass-Steel-Critchlow-Fligner test ( Table 3 ).
RESULTS
Saliva, sorbet, and juice pH values were 7.0, 3.8 and 4.9, respectively. Table 3 shows the roughness and colour ordinates baseline and final values. ΔE* values were 6.6 for saliva, 6.9 for sorbet, and 7.8 for juice. All specimens showed clinically significant pigmentation (ΔE>3.3) regardless of the solution. There was a significant ΔE* difference between saliva (p=0.005) and juice (p=0.002), and between juice and sorbet (p=0.019), but none between saliva and sorbet (p=0.401). No difference between juice and sorbet was observed for Δa*, but they differed significantly from saliva (p<0.001). Brightness (L*) changed significantly. Juice had the highest ΔE* (7.8) and ΔL* (7.7).
DISCUSSION
This study shows that aqaí might discolour RMGIC, although there is no change in roughness. Specimen standardizaron was used to manage bias. For instance, saliva/aqaí was removed from the freezer/refrigerator immediately before checking pH and the beginning of the protocol.
Solution | Composition | Manufacturer/Batch No. |
---|---|---|
Artificial saliva | Calcium (0.1169 g of calcium hydroxide/L of deionized water); 0.9 mM of phosphorus and potassium (0.1225 g potassium phosphate monobasic/L of deionized water); 20 mM TRIS buffer (2.4280 g TRIS buffer/L of deionized water) | Pharmaderm, Cascavel-PR, Brazil. |
Sorbet | Açai pulp, water, sugar, guar gum, carboxymethyl cellulose, tara gum, natural guarana extract, citric acid acidulant, natural guarana aroma identical, glucose, artificial dye amaranth and brilliant blue FCF, xanthan gum. | Polpa Norte, Japurá-PR, Brazil; 0136(TB) |
Juice | Medium açai pulp (100 g) blended with water (100 ml) | Polpa Norte, Japurá-PR, Brazil; 0430 |
Treatment | ||||
---|---|---|---|---|
Saliva | Sorbet | Juice | p values | |
Ra | 0.15 (0.10) A | 0.18 (0.10) A | 0.19 (0.10) A | 0.159 KW |
Ra' | 0.15 (0.15) A | 0.15 (0.10) A | 0.20 (0.18) A | 0.241 KW |
∆Ra | -0.0008 (0.13) A | - 0.03 (0.12) A | 0.00 (0.18) A | 0.660 KW |
L | 69.1 (2.1) A | 68.5 (2.3) A | 69.2 (1.7) A | 0.502 OwA |
L' | 62.5 (1.6) A | 61.7 (1.7) A | 61.4 (1.7) A | 0.073 KW |
∆L | -6.5 (1.3) A | -6.8 (1.7) AB | -7.7 (1.1) B | 0.003 OwA |
a* | - 1.7 (0.4) A | - 1.6 (0.3) A | - 1.5 (0.4) A | 0.367 OwA |
a'* | - 1,9 (0.2) A | - 1.1 (0.8) B | - 1.0 (0.7) B | <.001 KW |
∆a* | -0.3 (0.2) A | 0.5 (0.5) B | 0.5 (0.5) B | <.001 KW |
b* | 11.7 (1.2) A | 10.9 (1.1) A | 11.2 (1.0) A | 0.087 OwA |
b'* | 11.9 (1.1) A | 11.4 (0.8) A | 11.5 (1.1) A | 0.316 OwA |
∆b* | 0.2 (0.8) A | 0.5 (0.7) A | 0.2 (0.9) A | 0.229 KW |
∆E* | 6.6 (1.2) A | 6.9 (1.6) A | 7.8 (1.1) B | 0.002 OwA |
The three solution ΔE* valúes (>3.3) supported rejection of the nuil hypothesis. A?ai juice made the specimens darker with lower lightness coordinates (AL) 7 , and the positive Δa* indicates a shift towards red. These results suggests that the coloured compounds in the fruit and the liquid have stronger staining potential than those in the sorbet. Moreover, staining could be critical for other fruit colours. We speculate that staining was limited to the glaze applied to prevent surface irregularities 15 . The BIS-GMA (glaze) and HEMA (matrix and liquid content) 16 may have increased pigment transportation during the test. Glaze removal by brushing machine and a long-term protocol would show the possibility of body discolouration. The colour results for the saliva group might have been related to carboxymethyl-cellulose (a thickening agent), which can cause discolouration. Even composite resins showed noteworthy ΔE* valúes (7.8 to 10.6) when immersed in artificial CMC-based saliva 17 . The negative Δa* and AL valúes indicate a shift towards green colour and darker specimens, respectively.
A polyester strip provided a smooth surface 14 , which is why the samples were not polished 15 . The surface roughness variation (AR<0.00) was not significant (null hypothesis accepted). Despite the conflicting reports, experimental dissimilarities may explain disagreements 18 . The longer the time in coloured acid solution, the more colour would change. Scanning electron microscopy could provide information about the specimen surfaces. Inter-RMGIC comparison is still under investigation. The authors conclude that aqaí and saliva led to unacceptable RMGIC staining, but no significant change in its roughness.