Although it is plausible to consider the use of bioactive materials with the purpose of controlling the DEM-REM process, the isolated analysis of materials makes the clinical choice a difficult one. At the same time, the limitations of in vitro studies are relevant in terms of representing the complexity of the oral environment and other etiological factors of dental caries. Even in situ studies that seek to expose the substrate to an oral environment may overestimate the rates of progression of caries lesions [85]. Therefore, the present study sought to compile the existing bioactive restorative materials on the market, containing fluoride, calcium and/or phosphate, aiming at a critical analysis of methodological issues found in the clinical trials published to date.
The ranking of restorative materials used in clinical practice showed a better performance of GIC for the control of SC in permanent teeth, whereas for deciduous teeth, the best performance for the same outcome was demonstrated by the RMGIC. Therefore, among the bioactive materials, those with greater release of F ions demonstrated greater efficacy in controlling SC. The other restorative materials, such as compomer, gionomer and Activa, do not have high fluoride release or even adequate fluoride recharge [86]. Fluoride release and recharge characteristics depend on the matrix, filler type, as well as fluoride content and type in the material [20]. In addition to the composition itself and the setting mechanism, the need to use an adhesive system prevents the passage of ions from the restorative material to the tooth structure, making the REM process difficult [86]. This justifies the inferior clinical result for the SC outcome of compomer, gionomer and Activa, when compared to conventional restorative materials such as amalgam and resin composite [23,33,34,52,54,75].
When comparing restorative techniques with bioactive and conventional materials in the control of dental DEM, it was observed that the good oral hygiene of selected patients is a factor that can influence the results regarding the development of SC [58], as well as biofilm control, advice on diet and exposure to fluoride performed during patient follow-up in clinical evaluations [14,52,87]. This has gained theoretical-methodological support considering that, the absence of SC in posterior teeth, in RC and GIC restorations classes I and II, after different periods of clinical follow-up (1, 3, 4, 6 and 10 years) was attributed to the patients' good oral hygiene status [79]. Therefore, the risk of the formation of new lesions in these patients was considered low, regardless of the material used.
Regarding the specific protection of fluoride in individuals with a low or moderate risk of caries, the fluoride obtained from toothpaste and drinking water is sufficient to prevent the appearance of new lesions. In individuals at high risk of dental caries, the frequent decrease in pH hinders the action of sources with low concentration of fluoride, thus requiring additional sources, such as the restorative material itself [20,52]. Bioactive materials can, therefore, effectively contribute to the control of the DEM-REM process, attenuating the progression of caries lesions, implying a reduction in the frequency of SC. Among the bioactive materials evaluated, GIC and RMGIC have advantages over RC, such as lower technical sensitivity and demand for less clinical time, due to the easy handling and insertion into the cavity [48]. Thus, they can be considered as first-choice materials, the GIC for permanent dentition and RMGIC for deciduous dentition, in high-risk and difficult-to-treat patients, such as non-collaborating children and special patients, as well as individuals living in regions with difficult access to oral health care, using the ART technique [63,88].
The fact that the RMGIC is considered superior to the GIC in the deciduous dentition can be explained by the frequent GIC failures, especially in the proximal walls of the deciduous teeth. Access difficulty, small extension and lack of protection of the GIC can lead to porosity, as detected in clinical studies [67,74] and the consequent loss of the proximal wall. GIC restorations showed defects with the aspect of concavities in the proximal wall [63], which may predispose to the development of caries lesions due to biofilm retention. Therefore, the technical difficulty and the low mechanical property of the GIC become more evident in deciduous teeth, with the use of RMGIC being preferable in this case.
The longevity of a restoration is multifactorial, as it depends on the handling, the operator, the adhesion capacity of the material, the way it is applied, in addition to factors related to the patient [47]. Although amalgam remains a restorative material of choice in different countries, given the current evidence and evolution of restorative materials, it is necessary to implement preventive procedures, as well as select and develop materials with less invasive approaches [89,90]. Thus, although the amalgam with fluoride is ranked as one of the best materials for the control of SC in permanent teeth, at a follow-up of 2 years or more, this material was discontinued in the dental trade and is not currently available for clinical use. Also regarding the amalgam, especially in deciduous teeth, its use can lead to pulp exposure when the preparation principles are followed or even restoration loss, when these principles are not followed, making it difficult to retain the restoration. Therefore, its use in the deciduous dentition is not recommended.
Another point to be considered regarding clinical studies is the sample size. Usually, studies involving GIC and RMGIC showed the most representative number, which may impact the results, considering the variability found in patients. The fact that these materials have been on the market for a longer time than other bioactive restorative materials also influences the results, due to the greater number of clinical studies available in the literature. This can be observed in network analyses, where the nodes (circles) represent the sample size and the connections (lines) the number of studies.
SC was just one of the criteria observed in clinical studies and it can be observed that there is great difficulty for evaluators to attain a correct diagnosis and identification of the clinical characteristics of lesions at the margins of restorations. The presence of gaps without active lesions, discolorations and deterioration of the restoration margins can be mistakenly interpreted as early stages of SC, just as radiolucency in radiographic images may be indicative of residual caries or adhesive systems without radiopacity [91,92]. Therefore, the studies may have a greater underreporting of the therapeutic effects of bioactive materials in relation to those that are not bioactive [69], in addition to promoting unnecessary restorative reinterventions.
The selection of research participants can also influence the results obtained from SC cases [24,67]. In well-standardized studies, which select patients without systemic diseases, with good oral hygiene, who collaborate and have no bruxism and clenching, there is a great chance of similarity between different materials regarding clinical longevity. As the vast majority of randomized controlled clinical trials disclose this selectivity, they often demonstrate a situation that is so controlled it does not correspond to reality [93]. This fact indicates the need for reflection on the interference of controlled experimental designs. One knows how important they are; however, it is necessary to think that they do not necessarily reflect the patient’s reality. Thus, results may be overestimated or underestimated, considering that the environment (oral cavity) and the experimental period differ from the reality experienced by the patient. Hence, it is worth considering that there may be consequences of controlled experimental procedures when these results are translated into clinical practice.
In the qualitative analysis of the included studies, the blinding of the participants as well as of the evaluators attained an unsatisfactory result regarding the risk of bias. However, these are different materials, usually with different characteristics regarding color and texture, making it impossible to blind both the participants and the evaluators, which justifies this result. In some studies, a performance bias was observed by using different types of isolation of the operative field, both absolute and relative, between the control and comparison groups [30,31,57,61], or by selecting the group that received oral health instruction and oral hygiene procedures [28]. That also happens with the selection bias, for a study where the allocation to the control group occurred in one school and the comparison group in another, therefore, without randomization [28,62]. Few studies, as mentioned above, showed a risk of bias to be considered. The comprehensive search did not detect any publication bias. This confirms that the construction of the network analysis was robust, and our findings are not statistical artifacts.
Although there has been a large investment in the marketing of new products to control the DEM-REM process, preventing DEM and/or promoting REM, the restorative treatment alone does not control caries disease. It is necessary to identify the unbalanced risk factor exhibited by the patient and carry out an individualized control. However, the restorative material can be one more resource to be used in patients who are at high risk of dental caries.