A prospective and randomized clinical trial evaluating the effectiveness of ART restorations with high-viscosity glass-ionomer cement versus conventional restorations with resin composite in Class II cavities of permanent teeth: two-year follow-up

Abstract Objective To compare the effectiveness of ART restorations using High Viscosity Glass-ionomer cement (HVGIC) with conventional restorations using resin composite in Class II cavities of permanent teeth, in a 2-year follow-up. Methodology Seventy-seven restorations were made with each restorative material, Equia Fil-GC Corporation (ART restorations) and Z350-3M (conventional restoration), in 54 participants in this parallel and randomized clinical trial. Restorations were evaluated at 6 months, 1 and 2 years using the ART and the modified United States Public Health Service (USPHS) criteria. Chi-square test and Survival Analysis (p<0.05) were used for statistical analysis. Results The success rates for ART restorations were 98.7% (6 months) and 95.8% (1 year) for both criteria. At 2 years, success rate was 92% and 90.3% when scored by the modified USPHS and ART criteria (p=0.466), respectively. The success rates for conventional restorations were 100% (6 months), 98.7% (1 year) and 91.5% (2 years) for both assessment criteria. ART restorations presented a lower survival rate by the criterion of ART (83.7%) when compared to the modified USPHS criterion of (87.8%), after 2 years (p=0.051). The survival of conventional restorations was 90.7% for both evaluation criteria. Conclusion At the 2-years follow-up evaluation, no statistically significant difference was observed between the success rate of ART restorations with HVGIC compared to conventional restorations with resin composite in Class II cavities of permanent teeth.


Introduction
Currently, the main alternatives of direct restorative material to substitute dental amalgam are resin composite and polyalcenoate-based materials, with the glass-ionomer cements (GIC) being the most biomimetic ones. 1 In contrast to resin composite, the GIC presents a coefficient of linear thermal expansion similar to tooth structures and it releases fluoride, characterizing its anticariogenic property. 2 GICs emerged as the most suitable restorative materials in early studies on the impact of ART to oral health. 3 Today, ART is widely accepted by the international scientific community and used worldwide. 4 Although High Viscosity Glass-ionomer cement (HVGIC) is the material of choice for ART restorations, there is still room for improvements. Thus, some authors have proposed and tested additional retention in cavities restored with GICs to provide greater longevity to restorations in permanent teeth. 5,6 Further, encapsulating HVGICs led to in vitro increased flexural strength, 7 with possible positive influences to the restorative treatment.
According to a systematic review, 8 it cannot be suggested that resin composite has higher failure rates and risk for secondary caries than amalgam restorations due to the weak scientific evidence. Thus, with the Minamata Convention and the reduction in the use of mercury in several fields, including dentistry, resin composite restorations are considered viable alternatives to amalgam restorations. 9 Therefore, in studies looking for new restorative alternatives, resin composites must be considered control.
Mickenautsch 10 (2016) investigated the state of the art of direct restorations in posterior permanent teeth applying HVGICs. The author concluded that ART restorations showed similar clinical performance to amalgam restorations. Kielbassa, et al. 9 (2017) proposed that the promising HVGIC Equia Fil could be an option to dental amalgam in countries where health services do not cover resin composites or in cases of allergy to polymers.
Considering the lack of randomized and parallel clinical trials with high internal validity comparing resin composites and HVGICs, both in deciduous teeth and permanent teeth, it is difficult to indicate the superiority of a material, 11 especially considering the substitution for dental amalgam. Few studies have investigated the clinical performance of multiplesurface restorations using GICs and resin composites in permanent teeth. 12-14 Evaluating restorations performed with HVGICs under the ART approach would provide important data considering three aspects: testing a substitute for dental amalgam, the ART approach being a more socially available technique due to the non-use of electrical equipment, and ART might be an important approach in COVID era as it does not generate aerosols. 1,8,[15][16][17] 19 from which a mean DMFT-score was retrieved.
The exclusion criteria comprised the following: participants presenting mobile teeth, having paranormal occlusion, more than two multiple-surface cavities in permanent teeth and wearing orthodontic appliances.
The size of the cavity was classified as small, medium or large. 18 Only children or adolescents whose parents or the participant signed the Informed Consent form were included in the study. Stratified randomization was performed by subdividing class II cavities into two homogeneous groups. Cavity size and caries experience (DMFT) were the stratification variables, in that order.
For stratified randomization, due to the difficulty in obtaining the calculated number of class II cavities, the tooth was considered a sample unit. The participants were initially screened and in an Excel spreadsheet the screened teeth were listed with the indication of the cavity size and the patient's caries index. Teeth were initially ordered in this Excel spreadsheet by the caries index and divided into two conglomerates, the first with the lower DMFT values and the second with the higher values. After this division, in each of these clusters, the teeth were ordered by the size of the cavity and each half was divided into two parts, totaling four clusters namely: low DMFT-scores and small cavity size, low DMFT-scores and large cavity size; high DMFT-scores and small cavity size; and high DMFT-scores and large cavity size.
After this division into four groups, teeth were allocated to groups using the "random" function on Excel to ensure impartiality in the process of randomization and allocation. After this randomization, statistical tests were carried out to ensure that the factors used in the randomization were equally divided among the experimental groups. A T test for independent samples was used for the comparison between the caries index of the two experimental groups and a Mann-Whitney U test to compare the cavity sizes of the two groups (5%).

Training sessions
Prior to the beginning of the study, a single operator  Conventional restorations/resin composite protocol: The tooth to be restored was anesthetized and the operative field was isolated with rubber dam. The cavities were prepared using minimal invasive dentistry with # 245 or # 330 carbide burs at high speed. Carious dentin was removed with # 1, 2, or 3 round burs (KG Sorensen, Cotia, Brazil).

Results
In this study, the overall recalls at 2 years was 87%.
At the end of 2 years, two ART restorations received the score 9 according to ART criterion, being excluded from the analysis. Those restorations were also excluded from the evaluation with the modified USPHS criterion. Those restorations with code 9 (ART criterion) can be identified in Table 2. It is important to note that in this same evaluation period, another ART restoration classified with the score 6 according to the ART criterion was considered satisfactory as stated in the modified USPHS criterion (Tables 1 and 2).
Regardless of the evaluation criteria used, the success rates for conventional restoration with resin composite were the same. However, considering the success rates for ART restorations with HVGIC after 2 years, they were 90.3% when using the ART criterion *Chi-square test with linear trend. **The ideal and satisfactory scores = success; unsatisfactory = fail.    (6 months

Discussion
Both HVGIC and resin composite presented high success rates after 2 years (>90%). Although clinical success was similar within the assessed period of time, one might consider performing ART restorations since it has some advantages, among them the use of inexpensive hand instruments, only infected tooth tissue being removed, the employed material presenting chemical adhesion to the tooth substrates, and fluoride release. 19 In the present study, regarding ART restorations, the average annual failure rate was 6.3%, presenting a longevity of restorations higher than a metanalysis  that reported 78.2% 21 . The difference can in part be attributed to the material used, which was an encapsulated GIC, presenting improved mechanical properties compared with hand-mixed GICs commonly used in ART. 7 The present material also contains improved liquid and powder and the restoration surface is coated with nanofilled resin. The encapsulated GIC removes the negative effects of proportioning the powder/liquid ratio and diminishes the number of porous produced by hand-mixing. In addition to encapsulation, Equia Fil combines the main advantages of HVGIC, with a nanofilled, light cured coat, which protects the material in the initial setting phase and occludes surface cracks and porosity, increasing its wear resistance and toughness. 12,14,22,23 Other very important highlighted aspects for the present study are the time lapsed from the beginning of mixing and the initial removal of material excess (2.5 minutes) and matrix removal (5 minutes). Those are important to allow the initial material setting and hardening. 24 On the other hand, longer waiting periods are not desired due to difficult excess removal with possible occlusal interferences being left, a fact that would lead to early restoration fracture. The operator was aware and took into consideration bonding stability of HVGICs to dentin is not so strong in the early periods. 25 Moreover, a step that may also increase the longevity of multi-surface HVGIC restorations was the creation of retention grooves in proximal boxes, close to the amelodentinal junction, as noted by Barata, et al. 5 (2008) and Fernandes, et al. 6 (2019).
Regarding conventional restoration with resin composite, the average annual failure rate was 4.3%, being within the reported mean, which varies from 2 to 10% depending on the adhesive strategy used. [26][27][28] Generally, ART restorations are evaluated by ART criteria in most studies whilst the longevity of restorations are assessed by USPHS criterion. 15,29 It has been suggested that the ART criteria are more stringent than the USPHS criteria, since the marginal defect or wear exceeding 0.5 mm is considered to be a failure in the ART criteria and not for the USPHS criteria, which will consider failure only if dentin is exposed. 30,31 Moreover, the ART criteria of dental fracture considers failure even if the restoration remains intact, opposing the USPHS criterion that considers this scenario as success. Our results showed that regardless of the evaluation criteria used for restorations, the success rates were identical or similar, and there was no statistical difference between them according to the findings in the literature. [29][30][31] This was likely because the only two restorations with scores 3 and 4 (ART criteria) were also considered failures according to the USPHS criteria.
This study considered the use of the modified USPHS criteria adequate and comparable to the ART criteria. These criteria are relevant since they can assess marginal discoloration, color and surface texture, which are not measured by the ART criteria. 32 At the 6-month evaluation, differences were detected within the clinical acceptable parameters for both treatments (color, anatomical shape and surface texture). After 1 and 2 years, differences in anatomical shape and surface texture were detected between the restorative approaches ( Table 2). Differences in color were lost after the 6-month recall. Although there were differences between the restorations and their anatomical forms and superficial textures, during the 2 years of follow up, the quality of the restorations was not compromised. Since the anatomy of ART restorations is achieved by digital pressure, their anatomical shape would be a disadvantage over conventional restorations with resin composite.
Besides the resin composite being nanoparticulated, which ensures a high surface smoothness, the polishing of restorations performed also collaborate to a smoother surface texture, possibly explaining the differences found in surface texture between treatments.
The main reasons for failure in this study according to the ART criterion were: fracture of the restoration and/or tooth (9 restorations), painful symptomatology (5 restorations), one restoration failed due to excessive wear and one restoration failed due to a marginal defect of more than 0.5 mm. Only after 2 years one restoration fail due to secondary caries. The other fractures in the study probably occurred due to different intrinsic reasons of restorative materials, the technique employed or patient habits. 18,33 The results of the present study showed that the preventive effect of caries was similar for both materials. After 2 years, abscess and/or fistula were present in three ART restorations and two conventional restorations, suggesting a high level of efficacy after J Appl Oral Sci. 2021;29:e20200609 9/10 cement had been applied for the ART restorations, while calcium hydroxide solution and cement and GIC base were applied when necessary for conventional restorations with composite resin. Six other teeth with deep carious lesions such as these responded favorably to the protection of the dentin-pulp complex, maintaining pulp vitality and restorative success after 24 months.
One study reported 34 that after a three-year followup, the annual failure rates for resin composites and resin-modified GICs in deep cavities were 14.6% and 26.7%, respectively. Now, regarding the management of carious lesions, the annual failure rates for the restorations were 17.3% when the selective removal was performed and 13.1% when the total removal of the carious lesion was performed. Thus, in the present study, failure rates, especially for painful symptomatology due to pulp involvement, were considered low.
The present results do not follow the main reasons for the substitution of conventional restorations with resin composite observed in the literature, which are the development of secondary caries related to the adhesive interface and fracture of the restoration that is related to the mechanical properties of the material, as well as to the quantity and quality of the remaining dental structure. 35,36 Poor oral hygiene along with high rates of DMFT negatively impact the success of restorations in general. 37 Most publications do not report the oral health status of subjects included in the studies nor did they report whether a dental health program was implemented along with clinical treatment. 37 This may also have been one of the reasons for the observed restorative success of this study, because in addition to explaining how to clean their mouth and reinforcing the importance of preventive measures in each followup, the participants received a new kit for oral hygiene at each evaluation point.
By comparing the success rates of treatments (data raw) and the survival analysis values (from a probability curve), we observed that the survival analysis generally provided a lower restorative success for both approaches tested in the study. At this followup period, although the survival analysis presents lower values, they are close to the raw percentages detected. The authors are not aware if this trend would be more perceptible in future recalls. To our understanding, the survival analysis underestimates the real effectiveness of restorations in non-inferiority studies. [38][39][40] The results of the present study showed that the null hypothesis could not rejected, and there is no difference after 2 years on the effectiveness of these ART restorations with HVGIC compared to conventional restorations with resin composite.