Gloss retention of direct composites and corresponding CAD/CAM composite blocks

Abstract Objectives To compare gloss retention of four different resin composites with their corresponding CAD/CAM composite blocks. Materials and methods Four direct resin composites (Filtek Supreme XTE A2 Body (3M, USA), Tetric EvoCeram A2 (Ivoclar Vivadent, Liechtenstein), GrandioSO x‐tra A2 (VOCO, Germany), G‐aenial Universal A2 (GC, Japan)), and their corresponding CAD/CAM composite blocks were tested. A total of 288 samples were prepared and three different tests were performed: brushing, exposition to acidic fluoride gel and exposition to alcoholic solution. Gloss values were obtained by means of a glossmeter at T0 before aging and T60 after 1 h of aging. Results Mean gloss values ranged from 0.9 after brushing tests to 79.0 after the alcohol test witnessing a high gloss variability depending on the materials and the aging test. Statistical analysis by means of two‐way repeated measures ANOVA followed by Fisher's LSD post‐hoc test revealed significant differences between materials, storage media, and their interactions. Conclusion Gloss retention seems to be dependent on the composite type (direct or CAD/CAM block) and composite brand and varies in respect to the type of aging. CAD/CAM materials showed a higher resistance toward alcohol exposure.


| INTRODUCTION
Besides being the material of choice for direct restorations (Ardu & Krejci, 2006;Dietschi et al., 2012) composite resins in the form of prefabricated blocks are rapidly invading indirect CAD/CAM workflows. Most manufacturers have therefore launched the equivalent of their direct resin composites in the form of CAD/CAM blocks due to the multiple advantages of this type of materials, such as ease of repair, perfect compatibility with adhesive techniques, higher resilience compared to brittle ceramics, relatively lower cost, and sufficient mechanical properties for single-tooth restorations (Ardu et al., 2011;Dietschi et al., 2019;Jassé et al., 2013). Furthermore, the use of CAD/CAM composites for inlays, onlays, endocrowns, and even for anterior restorations allows for a decrease of general costs by cutting down dental laboratory expenses in the medium to the long term view (da Veiga et al., 2016;Dietschi et al., 2019), despite the initial investment of the acquisition and milling machines.
Composite CAD/CAM blocks are often made out of the same or very similar components as their respective light-cured direct restorative resin composites, with the difference that the curing step is performed by the manufacturers rather than clinically by the dentist.
This well-controlled professional curing under high pressure and high temperature leads to a higher conversion rate and to even, as claimed by some manufacturers, better clinically performing materials (Batalha-Silva et al., 2013). Enhanced properties are not limited to the mechanical aspect, but also to the optical and chemical stability, which reduces staining and increases gloss retention. More specifically, gloss retention is an important factor as it allows for a better esthetic appearance of composite restorations, especially in anterior area. A high lip line can cause, in fact, a reduced amount of saliva on the tooth surface, causing a progressive dull aspect of the restored tooth, especially if it is located between intact natural teeth (Lefever et al., 2012).
Exposure of resin composites in general, to acids, alcohol as well as brushing habits is known to alter their appearance on the long term, but no data exists on the comparison between the gloss retention of recently launched CAD/CAM composites and their corresponding traditional direct resin composites.
Therefore, the aim of this in vitro study was to compare the gloss retention of four restorative direct resin composites, and their respective indirect CAD/CAM blocks after challenging their surfaces with chemical and mechanical attacks. The null hypothesis was that mechanical and chemical agents were not able to significantly decrease surface gloss of direct composite resin materials as well as indirect CAD/CAM blocks.

| MATERIALS AND METHODS
A total of 288 samples were prepared, specifically 12 samples for each tested material and for each aging test (Table 1). The choice of materials was based on commonly used light-cured resin composites together with their respective CAD/CAM blocks. The manufacturers were also asked to confirm that the light-cured composites and their respective CAD/CAM blocks had a similar composition. A nano-filled composite, Filtek Supreme XTE, a hybrid composite with pre-

| DISCUSSION
The aim of this in vitro study was to compare the gloss retention of four restorative direct resin composites, and their respective indirect CAD/CAM blocks after challenging their surfaces with chemical and mechanical attacks. These common clinically relevant aging factors (Ardu et al., 2009;Goldstein & Lerner, 1991;Lefever et al., 2012Lefever et al., , 2014Neme et al., 2003;Tanoue et al., 2000), that is, brushing, exposition to acidic fluoride gel, and exposition to alcoholic solutions, may be present in routine diets and common oral hygiene habits and could highly influence the esthetic appearance of direct and indirect CAD/CAM composite restorations.
In our study design, we decided to go for manual polishing technique that was intentionally used in order to represent clinical reality.
This approach led to lower gloss values compared to results obtained by machine-polished samples with up to 4000 grit size paper (Lefever et al., 2012(Lefever et al., , 2014. The aging tests were standardized by either submerging the samples in the same medium for the alcohol and fluoride gelée tests and by using an automated brushing simulator. To avoid any subjective bias in the evaluation of the surface gloss, the use of a glossmeter that has the capacity to numerically report surface reflectance of a restricted area under standardized conditions was used.

T A B L E 2 Materials' initial and final values for the three aging tests and materials' rankings after aging tests, where A is the best and D is the worst
This allows to reduce confounding factors such as angle of the observer and illumination (Sheen Instruments Ltd, 2000) which was set at 60 for all measurements, in accordance with Da Costa et al. (2007).
The alcohol test was performed to evaluate the effect of possible softening by alcohol on the gloss of the composite surface. A solution of 75% ethanol was used according to previous studies (Ardu et al., 2009;Yap et al., 2003). Condon and Ferracane (1997) showed that aging through ethanol storage (75% ethanol aqueous solution, 37 C) produces an increase in subsequent wear only in composite materials that are under-cured, while no effect should be detected in In order to mimic acidic attacks, Elmex gelée was used as proposed by Ardu et al. (2009) due to the presence of highly concentrated aminofluoride. This kind of gels is widely and regularly employed in the field of caries prevention due to its presumed anticaries capacity. This formulation, when in contact with water, develops hydrofluoric acid and becomes quite aggressive against glass and ceramics as well as composite fillers (Wozniak et al., 1991) which leads to structural changes altering the gloss behavior. All tested materials were highly affected by the aminofluoride gel and no superiority of CAD/CAM blocks was found over the light-cured materials.
The results showed that the direct composite FS was slightly better in the FS/LU pair, while the CAD/CAM block CS was slightly better in the GU/CS pair, and that no difference existed in the two remaining pairs.
Concerning the brushing test, the same protocol was used as in previous studies (Hanasaki et al., 2018;Lefever et al., 2012Lefever et al., , 2014Wiegand et al., 2013) where the influence of each test parameter is widely discussed. Specifically, in this study, a medium abrasive (75 RDA) toothpaste was used, with a soft toothbrush and a standardized brushing force of 1.5 N by using a mechanical brushing device.
The direct light-cured materials containing nanofillers (Filtek Supreme) or fumed silica and silica glass dispersed into the matrix (G-aenial Universal) resisted better to this test than the hybrid composites and even than their relative CAD/CAM blocks. This rather surprising behavior could be partially due to a possible bias of this study as the time of renewing the toothpaste slurry was set to 5 min. Containing large amounts of rounded nano-particles, the detachment of such fillers could have left a more favorable topography than the hybrid materials (Tetric EvoCeram and GrandioSO x-tra), and detached nanofillers could have also acted as an additional polishing agent. In the particular cases of GrandiSO x-tra and Grandio CAD that showed the lowest gloss values in this test, a possible explanation could be related to a sub-optimal silanization of the filler particles which could lead to the so-called pothole effect after detachment, creating large surface cavitations resulting in huge gloss decrease. Our results are substantially in accordance with Ardu et al. (2020), and slightly different from Lefever et al. (2012Lefever et al. ( , 2014 who did similar research. These differences can be easily explained by the fact that we performed manual polishing of the samples with Sof-Lex discs while they did it by means of a polishing machine which allowed for higher initial gloss values.
Distilled water was used a negative control in the present study.
In particular, samples of all tested materials were dipped in distilled water for 1 h and measured before and after the immersion. Gloss values were almost identical and, obviously, no statistical differences were detected between values before and after aging in distilled water.
When judging on the clinical relevance of gloss variation in the three experiments, not only statistical differences have to be taken into account but also their clinical implications. According to Tessarin et al. (2018) ΔGU of 17.6 units is the limit of perceptibility of gloss variation, which is defined as the probability of at least 50% of observers to detect a gloss difference. Based on this criterium, alcohol did not affect human gloss perception with all materials tested, while Elmex gelée perceptibly affected all tested materials with the only exception of Cerasmart 270, and the brushing test caused clinically perceptible changes in all the tested materials.
Further in vitro and in vivo research with other direct composite as well as CAD/CAM blocks are needed in order to confirm these findings.
The null hypothesis that mechanical and chemical agents do not decrease surface gloss of direct composite resin materials as well as indirect CAD/CAM blocks was rejected.

| CONCLUSIONS
Gloss retention seems to be dependent on the composite type (direct or CAD/CAM block) and composite brand and varies in respect to the type of aging. CAD/CAM materials showed a higher resistance toward alcohol exposure. However, these results must be interpreted with caution as they are linked to the specificity of the experiment's setting.

ACKNOWLEDGMENTS
We would like to thank the manufacturers who provided generously the products tested in this study and Ms Isaline Rossier for the technical assistance during the experiments.