Clinical Effectiveness of Bulk-Fill and Conventional Resin Composite Restorations: Systematic Review and Meta-Analysis

The objective of this systematic review and meta-analysis was to determine the clinical effectiveness of bulk-fill and conventional resin in composite restorations. A bibliographic search was carried out until May 2020, in the biomedical databases Pubmed/MEDLINE, EMBASE, Scopus, CENTRAL and Web of Science. The study selection criteria were: randomized clinical trials, in English, with no time limit, with a follow-up greater than or equal to 6 months and that reported the clinical effects (absence of fractures, absence of discoloration or marginal staining, adequate adaptation marginal, absence of post-operative sensitivity, absence of secondary caries, adequate color stability and translucency, proper surface texture, proper anatomical form, adequate tooth integrity without wear, adequate restoration integrity, proper occlusion, absence of inflammation and adequate point of contact) of restorations made with conventional and bulk resins. The risk of bias of the study was analyzed using the Cochrane Manual of Systematic Reviews of Interventions. Sixteen articles were eligible and included in the study. The results indicated that there is no difference between restorations with conventional and bulk resins for the type of restoration, type of tooth restored and restoration technique used. However, further properly designed clinical studies are required in order to reach a better conclusion.


Introduction
Currently, the most common dental problem is dental caries, which is characterized as a bacterial infection causing damage to the tooth structure. For its treatment, dentists recommend removing carious dental tissue and filling the resulting cavity with the appropriate restorative materials. Currently, polymeric compounds are used as dental restorative materials due to their

Study Quality Assessment
The title and abstracts of all the articles identified by the electronic search were read and evaluated by four authors (M.L.-S., B.L., A.S.K., and S.P.) Disagreements between the reviewers were resolved by consensus with all the authors.
To evaluate the studies, a duplicate checklist was performed to extract the information of interest and change the data. Three reviewers (ML-S., A.S.K., and B.L.) independently assessed articles by name, author, year of publication, type of study, number of patients (male-female ratio), number of teeth restored, mean age and age range of patients, follow-up time, country where the study was conducted, study groups, number of patients and teeth per study group, type of restoration (class I, II, or V), type of tooth (incisor, canine, premolar, and molar), evaluation criteria, etching method, adhesive used, resin used, techniques used (incremental, bulk or bulk two-step), and the clinical parameters evaluated by each study. To resolve any discrepancies between the reviewers, they were discussed together with a third reviewer (H.I.A.-V.) to reach an agreement.

Assessment of the Risk of Bias of the Studies
The risk of bias assessment of each study was carried out by three authors (H.I.A.-V., S.P., and C.L.-L.) and was analyzed according to the Cochrane Handbook of Systematic Reviews of Interventions [13]. For the risk of bias, 7 items were analyzed and articles were grouped as being high, moderate, or low risk. An article was considered as low risk of bias, if all your items met the standards of the Cochrane Handbook of Systematic Reviews of Interventions; moderate risk of bias, if 1 or more

Analysis of Results
The data from each study were placed and analyzed in the RevMan 5.3 program (Cochrane Group, London, UK) using a relative risk (RR) measure and with a 95% confidence interval (CI).

Selection of Studies
A total of 1262 titles were identified from the database search carried out until May 2020. After removing the duplicates, 752 titles were thoroughly assessed based on the selection criteria. The fulltext assessment was carried out for 26 potentially eligible studies to identify 16 titles meeting the inclusion criteria. The reasons for exclusion is mentioned in the flowchart (Figure 1). The selected 16 articles were extensively reviewed for their content and their methodology for both qualitative and quantitative analyses ( Figure 1).
Only a few studies [18,21,23,24,28] mentioned using a rubber dam for moisture control during the clinical restorative procedure. Other included studies used cotton rolls and suction for isolation.

Synthesis of Results (Meta-Analysis)
Analysis of the clinical performance of conventional resins and bulk resins in restorations was based on the following 11 parameters: absence of fractures, absence of discoloration or marginal staining, adequate marginal adaptation, absence of post-operative sensitivity, absence of secondary caries, adequate color stability and translucency, proper surface texture, proper anatomical form, adequate tooth integrity (no wear), adequate restoration integrity, and proper occlusion (Appendix A). The other two clinical parameters, absence of inflammation and adequate point of contact, were not analyzed, since each of them was reported by only one included study (Al-Sheikh [24] and Balkaya et al. [15], respectively).
The subgroup analysis was performed based on the cavity form (class I/II and non-carious cervical lesions), type of dentition (primary or permanent), and tooth restoration technique (incremental or bulk or two-step bulk). The analyses showed that in the aspect of absence of fractures, absence of discoloration or marginal staining, adequate marginal adaptation, absence of secondary caries, adequate color stability and translucency, proper surface texture, proper anatomical form of the restoration, adequate integrity of the tooth without the presence of wear, adequate restoration integrity, and proper occlusion, there were no significant differences between conventional resins and bulk resins. The data were found to be homogeneous and around the line of no effect (Appendix A).
As for the absence of post-operative sensitivity, the analyses revealed that there were no significant differences when comparing a conventional resin with a bulk resin covered with a conventional resin (two-step bulk technique). However, regarding the clinical feature, there were significant differences

Synthesis of Results (Meta-Analysis)
Analysis of the clinical performance of conventional resins and bulk resins in restorations was based on the following 11 parameters: absence of fractures, absence of discoloration or marginal staining, adequate marginal adaptation, absence of post-operative sensitivity, absence of secondary caries, adequate color stability and translucency, proper surface texture, proper anatomical form, adequate tooth integrity (no wear), adequate restoration integrity, and proper occlusion (Appendix A). The other two clinical parameters, absence of inflammation and adequate point of contact, were not analyzed, since each of them was reported by only one included study (Al-Sheikh [24] and Balkaya et al. [15], respectively).
The subgroup analysis was performed based on the cavity form (class I/II and non-carious cervical lesions), type of dentition (primary or permanent), and tooth restoration technique (incremental or bulk or two-step bulk). The analyses showed that in the aspect of absence of fractures, absence of discoloration or marginal staining, adequate marginal adaptation, absence of secondary caries, adequate color stability and translucency, proper surface texture, proper anatomical form of the restoration, adequate integrity of the tooth without the presence of wear, adequate restoration integrity, and proper occlusion, there were no significant differences between conventional resins and bulk resins. The data were found to be homogeneous and around the line of no effect (Appendix A).
As for the absence of post-operative sensitivity, the analyses revealed that there were no significant differences when comparing a conventional resin with a bulk resin covered with a conventional resin (two-step bulk technique). However, regarding the clinical feature, there were significant differences between conventional resins and bulk resins in the type of restoration, type of tooth restored, and technique used (Appendix A, Figures A17-A19). The results were quite interesting, showing reduced or no post-operative sensitivity for NCCLs restored with composite resins rather than bulk ones (RR 1.11 95%CI [0.99, 1.23], p = 0.060, with an overall significant effect (RR 1.02 95%CI [1.00, 1.05], p = 0.05). A favorable effect of absence in post-operative sensitivity was also seen for cavities treated in permanent dentition (RR 1.03 95%CI [1.00, 1.06], p = 0.04) and with incremental technique for composite resins (RR 1.02 95%CI [1.00, 1.05], p = 0.05).

Discussion
In the present investigation, the null hypothesis was not rejected. The clinical effectiveness of bulk-fill resin is similar to conventional resin, regardless of the type of restoration (class I, II, or non-carious cervical lesions), the type of tooth restored (primary or permanent teeth), or the restoration technique used (incremental, bulk, or bulk two step).
The use of restorations based on light-curing composite resins has become widespread due to their adequate mechanical behavior and attractive aesthetic characteristics [30]. These restorations traditionally use a complex restoration technique, which is performed using a so-called "incremental technique" [31,32]. It is used for 2 reasons: (a) the depth of cure of these conventional resin materials is limited, preventing total polymerization in increments that are greater than 2 mm; and (b) an attempt is made to control the effects of material shrinkage when the polymerization reaction occurs [10,31,33,34]. Therefore, for deep or extensive preparations, several layers of the conventional resin material must be applied, which is technically challenging, consumes a lot clinical time, and also involves certain risks, such as the entrapment of air bubbles or contamination between layers. In response to these difficulties, a new generation of composite resins has appeared, called "bulk-fill" resins [10]. The increased translucency of these resins [10,35] due to their incorporation of more photoinitiator reagents [36] allows for deeper photopolymerization and permits insertion of the material into thick 4-5 mm increments, with uniform polymerization and degree of conversion. Furthermore, these resins can be clinically placed via 3 restoration techniques: the two-step bulk technique (using flowable bulk-fill covered with conventional resin material), the bulk technique with sonic activation (using flowable bulk-fill with sonic activation), and the bulk technique (using paste-like or regular bulk-fill) [10]. All of these factors are essential to obtain more satisfactory mechanical properties, and consequently to increase the longevity of the restorations [37,38]. Additionally, bulk-fill resins contain polymerization modulators that achieve low shrinkage and less stress on the bonded interface [35,36,39]. The insertion of thicker increments also contributes to reducing the incorporation of air voids, forming a more homogeneous restorative unit [35,36].
It is known that the longevity of the restorations is related to the clinical situation in the patient's oral cavity [21,26]. For this reason, there is a greater number of restoration failures in patients with parafunctional habits or temporomandibular disorders. such as bruxism [40]. Van Dijken et al. [17,19] reported a considerable number of failures caused by material or tooth fractures, most of which were in patients with bruxism.
The presence of secondary caries may be associated with the presence of marginal defects in a restoration [41] or with patients with a high risk of caries [42]. Van Dijken et al. [17,19] did not exclude patients with this condition and confirmed that failure caused by secondary caries was associated with patients with a high risk of caries; therefore, secondary caries may be related to biological failure rather than the restorative material used [40,42,43].
The depth [44] and the extension [45] of the cavity are factors that can influence post-operative sensitivity. A single study [28] radiographically confirmed that the restorations were performed in cavities 4-5 mm deep, while the other studies evaluating post-operative sensitivity did not describe the depth of the preparations or did so in more shallow cavities.
In the present review, significant differences were found between bulk resins and conventional resins in their post-operative sensitivity. However, most of the included studies reported either no occurrence of post-operative sensitivity in restored teeth [14][15][16][17][18][19][20]27,28] or no differences in post-operative sensitivity between bulk-fill and conventional resins [24,26,29]. One study argued [16] that the lack of post-operative sensitivity resulted from the placement of liners (e.g., calcium hydroxide liners or resin-modified glass ionomers) in deep and very deep cavities. Another study [28] observed post-operative sensitivity after 12 months in only one tooth of all those restored with bulk-fill with sonic activation, attributing this to the lack of calcium hydroxide liner in the deep cavity. Additionally, one study [18] reported post-operative sensitivity in a bulk-fill-restored tooth, which disappeared at 12-month recall.
Several studies [21,22,25] found that bulk-fill resin placement resulted in higher rates of post-operative sensitivity compared to conventional resins. These studies [22,25] preformed restorations of non-carious cervical lesions using bulk-fill resins and reported significant failure due to post-operative sensitivity within one year follow-up. It was argued that the high scores for post-operative sensitivity resulted from the limitations of the USPHS scale, which only allows the evaluation of either the absence or presence of post-operative sensitivity, without considering the intensity of the symptom. Restoration of NCCL is clinically challenging, as cavities located in the hypersensitive cervical region are more susceptible to contamination. Hence, using universal adhesives with proven antibacterial activity [46,47] and low cytotoxic effect [48] would seem appropriate. However, the clinical performance of bulk-fill resins placed in NCCLS was acceptable, yielding longer than 1-year follow-up periods.
The included studies mostly used universal adhesives in self-etching mode when placing bulk-fill resins. These adhesives are gaining popularity among clinicians, allowing for simplified procedures, however their dentin bonding potential can be enhanced by modifying the application method [49,50]. One study [21] using the self-etch adhesive Xeno III attributed the post-operative sensitivity to the adhesive used rather than to the bulk-fill resin material. A recent study showed no statistically significant relationship between post-operative sensitivity and the bonding system used [51]. Moreover, no association was found between the technique used to place the bulk-fill resin (incremental and bulk) and the depth of the cavity in terms of post-operative sensitivity [45]. A Cochrane review found inconsistent evidence regarding the use of linings and restoration failures, particularly regarding post-operative sensitivity [52].
This meta-analysis showed that there were no significant differences between conventional and bulk-fill resin compounds in terms of the type of restoration, the type of tooth restored, and the technique used. The results of this systematic review and meta-analysis are similar to those of Veloso et al. [53] and Boaro et al. [54]. Both studies reported that the clinical performance of conventional and bulk-fill resin compounds in direct posterior tooth restorations was similar, within a follow-up period of 12 to 72 months and up to 10 years, respectively. These clinical findings might be explained by corresponding mechanical properties, i.e., the shrinkage stress (for regular consistency materials), flexural strength, and fracture strength [54]. However, some of the chemical-physical properties (shrinkage, polymerization stress, cup deflection, and microhardness) of bulk-fill resin composites were found to be superior when compared with conventional composites, regardless of the viscosity or application technique [54]. As for the degree of conversion of bulk-fill materials with flowable consistency (used in two-step bulk technique), this was similar to conventional composite resins with thicknesses of up to 2 mm and greater than conventional composites with thicknesses greater than 2 mm [54].
Another study [55] testing wear and microhardness found that bulk-fill resin materials have comparable microhardness to conventional ones and present minimal change in surface roughness upon wear. This in turn could decrease bacterial adhesion to the surface of the restoration. However, more than the surface roughness, the filler particle size might play a role in bacterial adhesion, showing greater adhesion to resin materials with larger filler particles [56]. The hardness and wear resistance of the composite resins seem to be linked, yet it is difficult to assess resin material behaviour upon ageing [57].
Given that there are no reported clinical differences between restorations made of conventional resin materials and bulk-fill resin materials (in two-step or bulk techniques), these results seems promising, as most clinicians prefer to work with easy-to-use, clinically reliable bulk-fill resin materials, the placement of which occupies less chair-time in the dental office.
The current study has some limitations, such as the design of the clinical trial and the follow-up period, which could influence the results of the clinical trial. Randomized clinical trials are a critical method of clinically evaluating new materials and treatments, because these studies are standardized to achieve greater clinical credibility and reliability. However, a detailed description and similar methods should be used to allow a comparison. The clinical trials included in this study used different bulk-fill restorative materials with different etching techniques, which made it more difficult to compare them.
All of the studies included in this review used the modified USPHS criteria, however there were some differences between each one, resulting in a lack of standardization. The USPHS [58] criteria are the most common criteria for clinical evaluation of restorations. However, they have shown limited sensitivity and the categories may not fully reflect the clinical success of the restorations [59]. Clinical trials that have used other criteria tend to detect failure rates more than 4 times higher than those produced by the USPHS. However, this may be due not only to the increased sensitivity of these criteria, but also to a host of other factors, such as the fact that not all new systems are fully validated.
Likewise, randomization and allocation concealment are critical to the design of randomized clinical trials to avoid selection bias. Most of the included studies did not provide a complete description of these steps. Göstemeyer et al. [60] reviewed the design and validity of randomized clinical trials dealing with dental restorations and observed a high risk of bias, mainly in the domains of allocation concealment (93% selection bias) and blinding of participants and staff (99% performance bias) or blinding of outcome assessment (46% detection bias). Blinding of the operator and examiners may, in certain cases, be more difficult or even impossible to do, depending on the materials studied. However, allocation concealment can be implemented in all trials.
The follow-up periods observed in this study ranged from 6 months to 10 years. Differences in the efficacy of therapies can be measured only after several years because failure behavior can vary, and because one type of material may be more susceptible to long-term dental fractures and the other to secondary caries. Therefore, long observation periods are essential to observe all relevant effects and differences. However, maintaining a population of participants for an extended period is extremely difficult and attrition bias is very common [61].
For all of these reasons, the authors recommend that the results of this review should be interpreted with caution. Additional randomized clinical trials with better designs are needed.

Conclusions
In light of the current evidence, the clinical performance of conventional resins and bulk resins for carious lesion restorations is similar. However, properly designed clinical studies are required to avoid the biases observed in this study in order to reach a better conclusion.