Introduction

Endodontically treated teeth are weakened by the loss of strategic tooth structure through restorative procedures and caries, iatrogenic factors, and location and shape of the endodontic access rather than by the endodontic procedures [1, 2]. In this context, a restoration which restores function and preserves the remaining tooth structure against fracture, in addition to esthetic considerations, is obligatory after endodontic treatment. Clearly, the choice of appropriate restoration will depend mainly on remaining tooth structure, form of the endodontic access, and functional requirements [3].

In the case of extensive loss of tooth structure, it has been reported that coronal coverage significantly improve the survival rate of endodontically treated molar and premolar teeth [4, 5], while teeth with moderate remaining tooth structure may require more conservative approaches; such as resin composite restorations. Direct resin composite restorations are often used for endodontically treated teeth due to their bonding ability to tooth structure, which allows minimal cavity preparation, and might strengthen the tooth and offer an alternative restorative technique with a relatively low cost to cuspal coverage restorations [6, 7].

However, the problem of resin composites’ polymerization shrinkage which may result in interfacial deficiencies, cuspal movements, cracked cusps, microleakage, and secondary caries, continues to be a concern [8]. Especially for endodontically treated teeth, this stress distribution is higher in the tooth than within the restoration and tooth-restoration interface, and increases the risk of tooth fracture [9].

Several attempts at minimizing this problem has been suggested such as placement of stress absorbing intermediary layers including flowable composites in sandwich techniques [10–16] and incremental placement [17–22]. It is common in clinical practice to use flowable resin composites in sandwich technique for restoring large cavities. In this technique, flowable resin composite is used for the gingival portion of the restoration, and resin composite is used to complete the repair. The use of flowable resin composites has been advised to absorb polymerization shrinkage stresses due to their lower modulus of elasticity. Their low-viscosity, elasticity, and wettability allow them to fill irregular margins of proximal boxes [23] and improve marginal sealing of resin composites especially at the cervical level [24].

Incremental filling technique is another attempt to reduce polymerization shrinkage stress and its clinically undesired results. With this technique, the ratio of bonded surfaces to free surfaces (configuration factor) is minimized by reducing the contact area between resin and cavity walls. Unfortunately for the clinical procedure, applying multiple small increments and polymerization process of all increments make the restorative procedure time consuming and increases void generation and failure risk [25, 26]. Despite these attempts to improve the longevity of restorations, dentists still desire easier and quicker restorations.

In order to overcome the shortcomings of incremental filling technique with conventional composites, a new class of composites called “bulk-fill composites,” including low-viscosity (flowable) and high-viscosity (sculptable) material types, have been developed. The manufacturers of these newer resin composites claim that their composites can be efficiently photopolymerized at depths up to 4–5 mm in bulk and maintain low polymerization shrinkage stress at the same time.

While restoring large class II cavities, especially after endodontic treatment, it may be time-saving and practical for clinicians to complete the proximal boxes with the bulk-fill flowable composites and use conventional composites for the remaining cavity. A number of studies have been conducted to evaluate their degree of conversion [27, 28], depth of cure [29], and polymerization shrinkage [30–32], as well as their mechanical properties [33, 34]. There is no clinical evidence which compared the performance of new bulk-fill composites and placement techniques (open or closed sandwich) in endodontically treated teeth. The clinical evaluation of these bulk-fill composites is important as using these materials with open-sandwich technique will eliminate the disadvantages conventional technique which is slow and time-consuming.

The aim of this study was to investigate the clinical effectiveness of direct resin composite restorations placed with different placement techniques (incremental or bulk) with different flowable linings (conventional or bulk-fill flowable) used in open-sandwich technique, in endodontically treated class II cavities (mesio-occlusal or disto-occlusal), in a randomized controlled comparison. The null hypothesis tested was that restorations placed with different techniques and flowable linings would not show different clinical performances during 3-year period.

Materials and methods

Patient selection

The protocol and consent form for this clinical study were reviewed and approved by the Ondokuz Mayis University Human Ethics Committee (OMU-TAEK 2011/482) and registered to ClinicalTrials.gov (NCT02858947). During March–June 2011, patients attending the Restorative Dentistry clinic at Ondokuz Mayis University Faculty of Dentistry with at least two endodontic treatment requirements, caused by caries or fractures, irreversible pulpitis, or pulpal necrosis, were asked to participate in this follow-up study. All participants were informed about the study, advised of their rights, chance to deny the participation to the follow-up visits, and had signed the written informed consent prior to the first treatment. With this form, the patients accepted to participate to the study for 3 years (including follow-up visits).

Patients were excluded who needed indirect restorations due to significant loss of tooth structure, known allergies to product ingredients, had poor oral hygiene and a history of bruxism. To detect a difference between the restoration groups according to the retention rate of 30 %, with 90 % power and 5 % type one error rate, it is found that at least 47 teeth should be taken in each group. Permanent premolars and molars without any restorations were included in the study. Every tooth included in the study had neighboring teeth and were in occlusion to antagonistic teeth.

At the end of the patient-selection period, 37 patients (16 male, 21 female) aged between 19 and 41 years (mean 27) were found adequate for the current study.

Restorative procedure

The resin composites and adhesive system used in the current study are listed in Table 1. A total of 94 teeth (41 premolars, 53 molars) were endodontically treated by an expert operator. In the current study, endodontically treated class II cavities were chosen since these kinds of cavities are so common in clinical practice and it is important for clinicians to decide adequate technique and material for restoring these cavities. The enamel margins of the cavities were not beveled and cavity preparations were limited to removal of carious tissue.

Table 1 Materials used in the study
Full size table

After endodontic treatment, all tested teeth were temporarily restored with glass ionomer cement (Riva light cure, Southern Dental Industries-SDI, Australia) for 1 week. After 1 week, the temporary restorations were removed with diamond fissure burs (DIATECH, Swiss Dental, Heerbrugg, Switzerland) in a high-speed hand-piece underwater coolant. The average buccolingual width of each preparation was greater than one third of the distance between cusp tips. The teeth were isolated with cotton rolls and suction. Thirty-five percent phosphoric acid (Scotchbond, 3 M ESPE Dental Products, St. Paul, MN, USA) was applied for 30 s for the enamel margins and 15 s for dentine, then rinsed for 15 s with water, and dried with air-syringe. After acid-etching, Adper Single Bond 2 (3M ESPE Dental Products, St. Paul, MN, USA) was applied in two coats and light cured for 10 s, in accordance with manufacturer’s instructions.

By tossing a coin, a total of 94 mesio-occlusal or disto-occlusal restorations were placed on the permanent premolars and molars with either a conventional flowable resin composite lining (group 1), Aelite Flo (Bisco Inc. Schaumburg, IL, USA), or a bulk-fill flowable resin composite lining (group 2), x-tra base (VOCO, Cuxhaven, Germany). Aelite Flo used in the current study represented as a low-modulus microhybrid flowable composite specially developed with a patented long chain monomer which supplements the conventional Bis-GMA resin to provide a more elastic composite. This allows restorations to flex with the tooth for improved marginal integrity and overall restoration longevity. Low modulus of elasticity microscopically bends with the natural flexure of the tooth to prevent de-bonding and marginal breakdown [35]. The other tested flowable composite, x-tra base, is a bulk-fill composite which is one of the latest version of composites to simplify and shorten the restorative procedure.

In group 1, Aelite Flo, was applied in approximately 2 mm thick, light cured for 20 s and the remaining cavity was restored incrementally with a universal hybrid resin composite, GrandioSO (VOCO, Cuxhaven, Germany). The thickness of resin composite was not exceeding 2 mm and each increment was polymerized for 20 s. In group 2, x-tra base was applied in approximately 4 mm thick in bulk increments as needed to fill the cavity 2 mm short of the occlusal cavosurface and each increment was light cured for 20 s. The remaining occlusal part of the cavity was restored with a GrandioSO layer. GrandioSO was chosen because of its low polymerization shrinkage due to increased filler loading (71 % filler volume fraction) reported by a previous study [36]. The shade of the composite was selected according to the teeth to be restored. Metal matrices and wooden wedges were used for class II cavities. All cavities were restored with open-sandwich technique.

A LED device (Elipar S10, 3M ESPE Dental Products, St. Paul, MN, USA) with an irradiance of 1200 mW/cm2 was used for all polymerization process. The curing light intensity was measured with a radiometer (Curing Radiometer Model 100; Demetron Corp, USA).

After placement of restorations, the occlusion was checked with articulation paper. Fine grit diamond burs (DIATECH, Swiss Dental, Heerbrugg, Switzerland) and rubber cups (Edenta AG, AU SG, Switzerland) were used for finishing and polishing. All of the restorative procedures were carried out by the same operator. Blinding of the operator was not possible, because she knew the technique and type of lining material applied. This operator did not attend the evaluation process.

Evaluation

This study was double-blinded as neither the patients nor the evaluators were unaware of which flowable resin composite had been used. At the beginning of the study, two other examiners were calibrated by rating 20 high-resolution clinical photographs of posterior composite restorations that were representative of each score for each criterion. In the case of disagreement on a rating, both re-examined the restoration and arrived at a final joint decision in order to obtain only one score for each restoration. The intra-examiner Cohen’s kappa statistic was found high (0.95).

At baseline (after placement of restoration) and after 6 months, 1, 2, and 3 years, two calibrated examiners independently evaluated the restorations by Modified United States Public Health Service criteria (USPHS) [37] for the following characteristics: retention, anatomical form, marginal adaptation, color matching, marginal staining, surface texture, and secondary caries (Table 2). The resin composite restorations were evaluated with the aid of a dental explorer and an intraoral mirror and visual inspection. Bitewing radiographs were also taken at all recall time. The restorations were scored as follows: Alfa represented the ideal clinical situation, Bravo was clinically acceptable, and Charlie represented a clinically unacceptable situation. In case of disagreement between examiners, restorations were reevaluated and a consensus has been reached.

Table 2 Modified US Public Health Service Evaluation criteria [22]
Full size table

Statistical evaluation

The comparison of restorations with Aelite Flo and x-tra base linings for each category was performed with the Pearson chi square test. Within group differences of the materials at the baseline and after each recall time (6 months, 1, 2, and 3 years) was evaluated by Cochran Q test (p < 0.05). All data were analyzed by means of SPSS 11.5 for Windows (SPSS Inc., Chicago, IL, USA).

Results

The distribution of the number of restorations at baseline is given in Table 3. One restoration from each group did not match the shade and translucency of adjacent tooth structure at baseline and gave Bravo scores in terms of color match.

Table 3 Distribution of the number of restorations at baseline
Full size table

The recall rate was 100 % at 6-month and 1-year recalls. At 1-year recall, one restoration from x-tra base group could not be evaluated because that tooth was prepared to be treated with a zirconia crown. There were no other restorations lost and all other evaluated criteria were scored as Alfa for all restorations during 1-year period. There was no statistically significant difference between different groups (p > 0.05).

Twenty-six patients (14 male, 15 female) attended the 2-year recall, and a total of 72 restorations (36 for Aelite Flo, 36 for x-tra base) were available for evaluation. Only one restoration with x-tra base lining received Bravo score in terms of color match. Two restorations with Aelite Flo lining and one restoration with x-tra base lining showed gritty surface texture and scored as Bravo.

At 3-year recall, a total of 66 restorations, 33 restorations from each group, were available for evaluation. The same restoration’s color match from x-tra base group scored as Bravo during 3-year period. The surface texture of two restorations from each group received Bravo scores.

Table 4 presents the data of clinical evaluation of the materials used in this study over 3-year period. No statistically significant differences between the two groups were found for all evaluated criteria (p > 0.05). No secondary caries has been detected during evaluation periods.

Table 4 Clinical evaluation of the materials used in this study over 3 years
Full size table

Discussion

This study compared the clinical effectiveness of class II direct resin composite restorations placed with different placement techniques and different flowable linings, in endodontically treated teeth. The null hypothesis was accepted as resin composite restorations with different placement techniques and flowable composite linings resulted in statistically similar clinical parameters after 3 years.

Open-sandwich technique has been first described in 1977. Conventional glass ionomer cements (GIC) were used as lining materials in the early version of this technique [38]. Later on, resin-modified GIC and flowable composites were also advised to be used as lining materials [39]. In this clinical study, open-sandwich technique with flowable composite lining has been employed to test manufacturers’ claims. This technique has also been found to cause less microleakage and need less operator skill than closed-sandwich technique in previous studies [12, 24].

The use of a flowable composite liner as a flexible intermediate layer has been suggested for relieving the stress caused by polymerization shrinkage [13]. Using flowable composite liners has also been advised to exhibit superior contact with the floors and walls of cavity preparations [23]. Consistent with the previous data, the findings of this study showed that resin composite restorations with both of the tested flowable composite linings showed clinically acceptable performance with no secondary caries or restoration fracture at the end of the 3-year period. In contrast with our findings, some in vitro studies indicated unsuccessful cuspal deflection [19] and microleakage [14, 15] results of class II resin composite restorations with flowable linings than that without a flowable liner. These conflicting results may be due to the different materials and methodology used in these studies.

Several studies have been conducted to test the performance of bulk-fill composites. Depth of cure (DC) and polymerization shrinkage of these newer materials are some of the most common concerns about them. Par et al. [28] reported clinically acceptable DC of x-tra base at depths up to 4 mm. Similarly, Marovic et al. [31] also reported reduced shrinkage forces and high level of DC of x-tra base compared to a conventional flowable composite (EsthetX flow). The increased translucency of x-tra base due to its enlarged filler size might be a responsible factor of its high DC [33]. It is known that bigger particles decrease the filler-matrix interface and so allows more light transmission through the composite [34]. The high filler content of x-tra base (61 % volume) might be another reason for its low polymerization shrinkage because of the fact that the volume occupied by organic matrix and, therefore, the number of reactive methacrylate groups decreases [32]. Increased filler rate has also been shown to improve mechanical properties such as compressive strength, hardness, and resistance to softening [27]. The clinically good performance obtained with x-tra base may be explained by this way.

One of the main goals of adhesive dentistry is to obtain a tight interfacial adaptation, which reduces the risk of microleakage, secondary caries, and postoperative sensitivity. Incremental filling techniques have often been indicated to decrease the effects of shrinkage and stress generated at the adhesive interface and improves the marginal adaptation [40]. However, the effectiveness of incremental filling technique is controversial. Versluis et al. [9] showed that additional increments by incremental filling technique could produce higher shrinkage stresses at the adhesive interface and increase the cuspal deformation of the weakened cusps. Campodonico et al. [20] also reported no obvious advantage of incremental technique compared to bulk-filling technique in terms of cuspal deflection. In addition, incremental filling technique is a time-consuming procedure, may increase the contamination risk and include voids in the restoration [8]. In contrast, a recent in vitro study conducted by Bicalho et al. [21] reported that the negative effects of residual shrinkage stresses could be minimized by using a low post-gel shrinkage composite and increments that are not exceeding 2-mm thickness. Reduced cuspal deflection [18, 22] and higher resin–dentin micro-tensile bond strength in large cavities [41, 42] with incremental filling technique compared to the bulk filling technique have also been reported by previous studies.

In the current study, restorations placed with incremental technique and bulk-fill technique showed similar clinical performance during 3-year period. This clinically good performance of bulk-fill technique may be attributed to the reduced remaining cusp length after filling the cavity with x-tra base within 2 mm of the palatal cusp. It has been reported that the deformation of teeth is a cubed power of the cusp length [43, 44], and to reduce the deformation, cusp length should be reduced [16]. Similar with our results, the only study in the literature which evaluated the effectiveness of bulk fill flowable composite in a randomized controlled clinical study over 3 years reported that bulk-fill technique showed comparable clinical performance with incremental technique [8]. However, that study was conducted on vital teeth without endodontic treatment, so direct comparison of their results and the results of current study would be improper. Further clinical studies with bulk-fill composites are needed to confirm the reproducibility of the current study’s findings.

Conclusion

Within the limitations of the present study, it can be concluded that bulk-filling technique with x-tra base lining showed clinically acceptable performance comparable to the incremental technique with a conventional flowable lining over 3-year period.