Structural Integrity Evaluation of Large MOD Restorations Fabricated With a Bulk-Fill and a CAD/CAM Resin Composite Material
To evaluate the effect of two composite restorative techniques (direct bulk fill vs indirect CAD/CAM) on the fracture resistance and mode of fracture of extended mesio-occlusal-distal (MOD) cavity preparations. Fifty-one sound human mandibular third molars were divided into three groups (n=17). Extended bucco-lingual MOD cavities were prepared. Teeth in group 1 were restored with a bulk-fill resin composite (Filtek Bulk-Fill Posterior Restorative), teeth in group 2 were restored with composite computer-aided design/computer-aided manufacturing (CAD/CAM) inlays (Lava Ultimate), and teeth in group 3 served as control and remained intact. All specimens were submitted to thermocycling, and a fracture resistance test was performed using a Universal testing machine (0.5 mm/min). Mode of fracture was classified into five types. One-way analysis of variance and the Duncan test were used to analyze the fracture load data at a significance level of α = 0.05. A chi-square test was used for the analysis of fracture mode between the restorative groups. Statistical analysis showed significant differences in fracture resistance among the experimental groups. The teeth restored with the bulk-fill composite exhibited lower fracture resistance (1285.3±655.0 N) when compared to the teeth restored with the composite CAD/CAM inlays (1869.8±529.4 N) (p<0.05). Mode of fracture showed the same distribution between the restorative groups. Although both types of restorations failed at loads larger than those found in the oral cavity, the CAD/CAM composite inlays increased the fracture resistance of teeth with large MOD cavities when compared to direct bulk-fill composite restorations. The majority of fracture types were intraorally repairable for both restorative techniques.SUMMARY
Aims:
Methods:
Results:
Conclusions:
INTRODUCTION
Contraction stresses that develop during placement of resin composite (RC) materials due to polymerization shrinkage are a major drawback in contemporary restorative dentistry.1 Although RC restorative materials are commonly used in large and deep cavities, they often result in inconclusive success. Polymerization shrinkage of composite materials compromises the sustainability of the restoration since it may lead to decreased dentin bond strength, cuspal deflection, internal gap formation, and postoperative sensitivity.1,2 As a matter of fact, incremental buildup of multiple thin layers is required mainly to ensure curing and to potentially reduce the consequences of shrinkage stress.
Incremental filling techniques are considered clinically sensitive and time-consuming procedures. Manufacturers have recently introduced a novel type of RC, namely, the bulk-fill composite. This type of RC is marketed in either high or low viscosity, and its main advantage is that it can be placed in 4- to 5-mm-thick increments, thus eliminating the time-consuming element of RC placement. Furthermore, bulk-fill RC exhibits lower volumetric shrinkage, resulting in lower contraction stresses.3 In order to achieve sufficient polymerization depth, the fillers of bulk-fill composites were modified. In particular, there is an increase in filler size (>20 μm) and/or a reduction in filler load, mainly in the low viscosity bulk-fill resin-based composites.4 Furthermore, novel photoinitiators (ie, Ivocerin for Tetric EvoCeram Bulk-Fill) and different combinations of high-molecular-weight monomers aim to contribute to a desirable polymerization reaction.5
With regard to their performance, research has shown that bulk-fill RCs present similar and sometimes superior properties compared to conventional RC materials.3,6,7 As far as their clinical effectiveness is concerned, two randomized clinical studies showed that low-viscosity bulk-fill RC materials placed with a 4-mm bulk-fill technique presented good clinical effectiveness during a three- and a five-year follow-up, showing slightly better but not statistically significant durability compared to the conventional 2-mm layering technique.8,9
In an attempt to overcome the shortcomings of polymerization shrinkage and especially in large cavities, the use of indirect restorations is often recommended, where the contraction stresses are only limited in a small surface occupied by the luting cement between the tooth and the restoration.10,11 Inlays/onlays can be either laboratory (indirect technique) or chairside fabricated using computer-aided design/computer-aided manufacturing (CAD/CAM) technology (semidirect technique).12 The materials of choice for this type of restoration are either ceramic or RC. Recently, there has been an evolution of CAD/CAM resin-based materials. Various changes in their polymerization methods (high pressure, high temperature) as well as in their structure (glass ceramic networks) have improved their physical properties in comparison to their ceramic counterparts.13,14 Due to the complexity of their structure, different names have been used for this group of materials by their manufacturers, such as resin nanoceramics, hybrid ceramics, ceramic resins, or glass ceramics in a resin interpenetrating matrix.15
Today, composite CAD/CAM materials appear with enhanced mechanical properties, elastic modulus near that of dentin, and optimal stress absorbance in comparison to their laboratory counterparts. Furthermore, CAD/CAM allows construction of materials with low brittleness, small marginal crack percentages, safe intraoral reparability, and potential thickness up to 0.2 mm.16,17
Resin-based CAD/CAM materials have been investigated mainly in the form of inlays, onlays, and crowns with tooth preparation designs according to textbook traditional designs.18 It has been found that they present satisfactory behavior in terms of flexural strength and flexural modulus under occlusal stress conditions.19 Batalha-Silva and others20 concluded that CAD/CAM composite inlays increased the fatigue resistance and decreased the crack propensity of large mesio-occlusal-distal (MOD) restorations when compared to direct restorations. Moreover, Ender and others21 found that self-adhesive–luted CAD/CAM composite inlays showed similar marginal adaptation and fracture strength to glass-ceramic inlays.
There is limited literature regarding the properties of these two restorative modalities. Additionally, there is no evidence of their performance when a substantial amount of tooth structure is missing. It was therefore of interest to investigate the limitations of these materials when the size of the cavity is excessive. The purpose of this study was to evaluate the fracture resistance and mode of fracture of large MOD restorations using a highly viscous bulk-fill resin-based composite and a resin-based CAD/CAM material.
The null hypothesis was twofold. First, there will be no statistically significant difference between the experimental groups of the study regarding their fracture resistance. Second, there will be no statistically significant difference in the fracture mode between the two tested restorative techniques.
METHODS AND MATERIALS
Materials
A highly viscous bulk-fill resin-based composite (Filtek Bulk-Fill Posterior Restorative, 3M ESPE, St Paul, MN, USA) and a prepolymerized composite CAD/CAM material (Lava Ultimate, 3M ESPE) were investigated in this study. The shade of the composite materials was A3. The technical characteristics of the materials are presented in Table 1.
Preparation of Tooth Specimens
Fifty-one sound human third mandibular molars freshly extracted for orthodontic reasons were selected for this study. The teeth presented similar bucco-lingual and mesio-distal dimensions (≤10% difference) and were randomly divided into three groups of 17 teeth each. The teeth were examined for cracks or other structural deficiencies under an optical microscope (20× magnification). Calculus deposits and soft tissues were removed with a hand scaler (13S/14S modified Gracey, Hu-Friedy Manufacturing Co, Chicago, IL, USA) and an ultrasonic scaler (EMS SA, Nyon, Switzerland). The teeth were stored in 0.2% thymol solution (renewed every five days) for no longer than three months before the onset of the experiment. Before the experiment, the teeth were mounted in acrylic bases, covering each root up to 3 mm below the cemento-enamel junction with acrylic resin (Palapress, Heraeus Kulzer GmbH, Dormagen, Germany).
Restorative Procedures
In group 1 (Bulk-Fill [BF]), MOD cavities were prepared utilizing a high-speed hand piece (Allegra TE-95, W&H GmbH, Bürmoos, Austria) under water cooling using diamond burs (959 KRD, 8959 KR Komet Dental, Gebr. Brasseler GmbH & Co, KG, Lemgo, Germany), which were replaced every five preparations. The hand piece was adapted to a parallelometer in order to ensure repeatability of preparations as far as the cavity dimensions were concerned. Slot-type MOD cavities were prepared with rounded internal angles, diverging buccal and lingual walls (5° to 10°), 5-mm bucco-lingual width, and 4-mm occlusal depth (Figure 1). Cavities were restored using Filtek Bulk-Fill Posterior Restorative (3M ESPE); etching of the cavity walls was performed for 30 seconds in enamel and 15 seconds in dentin with 35% phosphoric acid (Scotchbond Universal Etchant, 3M ESPE) and abundant rinsing and air-drying. Subsequently, the bonding agent (Single Bond Universal Adhesive, 3M ESPE) was applied using an applicator tip and gentle implementation for 10 seconds. The bonding agent was then photopolymerized for 20 seconds using an LED light-curing unit (Elipar S10, 3M ESPE) at 1200 mW/cm2 with standard curing mode, according to the manufacturer's instructions. The bulk-fill composite was placed into the cavities in one increment, and the occlusal surface was formed using prefabricated occlusal matrices (Occlu-Print, Hager & Werken GmbH & Co KG, Duisburg, Germany) in order to have a repeatable pattern and to prevent an oxygen inhibition layer during curing. Each surface (mesial, occlusal, and distal) of the restorations was light cured for 20 seconds according to the manufacturer's instructions. Finishing and polishing of the restorations was performed with a polishing disk system (Sof-Lex finishing and polishing disks, 3M ESPE).
Citation: Operative Dentistry 44, 3; 10.2341/18-013-L
In group 2 (Lava Ultimate [LV]), cavity preparations followed the same method as described above for group 1. The teeth were restored with inlay restorations that were fabricated using the Cerec 3 CAD/CAM system, CEREC 3D, version 3.85, Sirona Dental Systems, GmbH, Bensheim, Germany) with an average thickness of 3.5 mm at the central groove. A resin-based CAD/CAM material (Lava Ultimate Restorative for CEREC, 3M ESPE) was used for the preparation of the inlays. Restorations were mechanically polished using the same polishing disk system as in group 1. Internal surface treatment of the restorations included air abrasion (Aquacare Twin Dental air abrasion unit, Velopex Int, Medivance Instruments Ltd, London, UK) with 29-μm aluminum oxide particles at 0.2-MPa air pressure and was followed by 35% phosphoric acid cleaning. The internal surfaces were then rinsed for 30 seconds, air-dried, and silanated with Monobond S (Ivoclar Vivadent AG, Schaan, Liechtenstein) using a microbrush; allowed to react for 60 seconds, and dispersed with a strong stream of air. Tooth preparations were thoroughly cleaned and air-dried before cementation. The luting material was a self-adhesive resin cement (RelyX Unicem, 3M ESPE). Photopolymerization was performed initially for three seconds, and after careful removal of the cement excess, each surface was light cured for 20 seconds in accordance with the manufacturer's instructions. Finally, finishing and polishing of the restorations was performed with the polishing disk system (Sof-Lex finishing and polishing disks, 3M ESPE).
In group 3 (intact [INT]), which was the control group of the study, the teeth remained intact.
Artificial Aging
The specimens were submitted to thermal stress in a thermal cycling machine (10,000 cycles in a sequence of 37°C, 5°C, 37°C, and 55°C) with a dwell time of 10 seconds and a transport time of five seconds. Following thermocycling, all specimens were stored in distilled water at 37 ± 1°C for 24 hours before fracture testing.
Fracture Resistance Test and Fracture Mode Distribution
Each specimen was loaded axially on their occlusal surface at a crosshead speed of 0.5 mm/min in a universal testing machine (Testometric M350-10KN, Linkoln Close, Rochdale, UK). In order to keep the specimen firmly in place, a customized stainless-steel device was made and placed on the testing machine. A plunger with a steel ball (6-mm diameter) that made a tripod contact with the cusps was used to transmit the compressive force until fracture occurred (Figure 2).
Citation: Operative Dentistry 44, 3; 10.2341/18-013-L
The mode of fracture of the restorations was detected after failure using a classification previously described by Burke and others.22 According to this classification, the specimens were assigned to the following categories based on the pattern of crown failure using standardized photography:
Type I - fractures involving minimal destruction of tooth structure
Type II - fractures involving one cusp and intact restoration
Type III - fractures involving at least one cusp and up to one-half of restoration
Type IV - fractures involving at least one cusp and more than one-half of restoration
Type V - severe fractures involving tooth structure completely and/or longitudinal fracture that may require extraction of the tooth.
Statistical Analysis
The data were statistically analyzed using SPSS Statistics 20.0 software (IBM Corp, Chicago, IL, USA). Normality and homogeneity of the data were checked using the Kolmogorov-Smirnov test (p=0.937 for BF, p=0.967 for LV, and p=0.849 for INT) and the Levene test (p=0.311), respectively. One-way analysis of variance and the Duncan test were used to compare the mean fracture load among the experimental groups. The level of statistical significance was preset at α = 0.05. In addition, a chi-square test was used for analysis of fracture mode between the restorative groups. The sample size (n=14) was calculated considering 80% power and a significance level of 5%.
RESULTS
Three specimens (two from group 3 and one from group 2) were rejected due to wrong fixation in the acrylic resin bases. Further, the box-plot method revealed one outlier in the control group (Figure 3). Therefore, it was excluded from the analysis to avoid any erroneous results. Table 2 shows the frequency of the remaining specimens, which were ultimately used throughout the analysis. It must be noted that in determining the group sample sizes, a rule of thumb was followed that suggests that in experimental research with tight experimental controls, successful research is possible with samples as small as 10 to 20.23 Further, according to the Levene test, these unequal sample sizes did not produce unequal variances. Therefore, SPSS was set up to deal with the problem of unequal sample sizes and unequal variances by using the Duncan post hoc test to compare the mean fracture load among the experimental groups.
Citation: Operative Dentistry 44, 3; 10.2341/18-013-L
Fracture Resistance Measurements
Mean values and standard deviations of fracture load (N) of the experimental groups of the study are illustrated in Figure 4. The bulk-fill composite restorations presented the lowest fracture resistance (1285.3±655.0 N) (p<0.05) followed by CAD/CAM composite restorations (1869.8±529.4 N), while the control group exhibited the highest fracture resistance (3198.0±490.7 N) (p<0.05). One-way analysis of variance and the Duncan test revealed that there was a statistically significant difference of the fracture resistance between the three groups (p<0.001). More specifically, there was a statistically significant difference between the restored groups (groups 1 and 2) as well as the restored and the intact teeth (group 3) (p<0.05).
Mode of Fracture Analysis
The mode of fracture of the specimens was reported immediately after the fracture resistance test, and the distribution is shown in Table 3. Representative photographs of each type of fracture are illustrated in Figure 5a-e. Application of the chi-square homogeneity test revealed that in all cases of fracture types, the distributions were the same between the two restorative groups.
Citation: Operative Dentistry 44, 3; 10.2341/18-013-L
DISCUSSION
This study evaluated the fracture resistance and mode of fracture of posterior teeth with cavity preparation designs with an extensive isthmus restored with two restorative techniques—a direct RC and a chairside CAD/CAM—in an attempt to assess the most appropriate in terms of maximizing the structural integrity of the restored teeth. Many times, the clinician is faced with the dilemma of whether to restore a cavity with a wide isthmus with an intracoronal restoration or to proceed with traditional cusp protection protocols with partial or full-coverage crowns. The introduction of new materials and the evolution in adhesive technology can potentially give more conservative options for the restoration of these teeth. Two modern materials were tested—a direct bulk-fill composite and a hybrid RC CAD/CAM—following the respective adhesive/cementation protocols as suggested by the manufacturers. It was not the intention to assess the effect of different adhesive protocols in the fracture resistance of the restored teeth, only the effect of the two different restorative approaches to the restoration of teeth with extensive cavity designs.
In the current investigation, the restorations were subjected to a static loading test until failure occurred after artificially aged via thermocycling. The type of load exerted was in the form of a compressive force perpendicular to the occlusal surface of the restored teeth and parallel to their longitudinal axis. A tripodic contact with the buccal and lingual cusps was achieved through a 6-mm-diameter sphere made of stainless steel.24-26 Regarding aging, the specimens were artificially aged via thermocycling and water storage. These procedures can compromise the mechanical properties of a material.26 These changes are attributed to water sorption, polymer network expansion, and frictional reduction between polymer chains and may affect the clinical performance of the restoration.28 Gale and Darvell29 reported that a sequence of 35°C, 15°C, 35°C, and 45°C with 28, two, 28, and two seconds and 10,000 cycles is equivalent to one-year of oral function. In the present study, 10,000 cycles were performed, corresponding to approximately 12 months of oral function. Additionally, a sequence of 37°C, 5°C, 37°C, and 55°C was selected with a transfer time of five seconds and a dwell time of 10 seconds, which replicates the abrupt changes in temperature occurring in the oral cavity avoiding excessive forces due to long immersion (ie, 28 seconds).
In the oral environment, restorations may be loaded during their lifetime up to 107 cycles,30 and during physiological function of the stomatognathic system, the stresses applied to posterior teeth may reach up to 300 N (usually 50 to 60 N) for a long duration of chewing cycles.31 Furthermore, in terms of in vivo loading, the masticatory cycle consists of a combination of vertical and lateral forces, subjecting the restoration to a variety of off-axis loading forces.32 In effect, dental restorations usually fail as a result of many loading cycles or from an accumulation of damage from stress and water rather than during a single application of a high chewing load.33 However, the intention of this investigation was to give a primary indication as to whether the proposed restorations would reinforce the structural integrity of the restored teeth.
In the present study, there was a significant difference in fracture resistance between the restorative groups (BF and LV). A possible explanation may be an increased crack formation in dental structures during the polymerization of direct restorations, which makes them more susceptible to fractures. Silva and others20 reported that large direct conventional composite restorations exhibited significantly higher crack propensity (47%) as a result of polymerization shrinkage stresses compared to CAD/CAM composite inlays (7%). Furthermore, one important advantage of composite inlays is that the polymerization of the composite materials takes place before the clinical procedure, and as a result there are no consequences of polymerization shrinkage stresses on tooth structure. This may be the explanation of the higher fracture resistance of the LV group compared to the BF group in the current investigation. Increased fracture resistance of LV restorations may also be attributed to high content of fillers, standardized production conditions of CAD/CAM materials, and similar flexural modulus of the material and adhesive cement, meaning that an increased breaking force is needed to create cracks that could cause fractures.34,35
Additionally, physical and mechanical properties of restorative materials, such as fracture toughness, modulus of elasticity, creep, hardness, and polymerization shrinkage, are important for the fracture resistance of a restoration.6 The modulus of elasticity of a material is one of the key factors determining the clinical performance of restorations in the testing of chewing forces. High-modulus materials tend to accumulate stresses in their mass, while materials with low modulus of elasticity tend to absorb the forces transferring them to the surrounding dental structures.36 When occlusal forces are applied to materials with high modulus of elasticity, tensile forces develop at the tooth/restoration interface just below the force application point.37 The elastic moduli of dentin (E=17.6 GPa), bulk-fill composite (E=15.2 GPa), and CAD/CAM composite (E=14.5 GPa) used in this study were similar.38 Consequently, the difference in the behavior of the materials in fracture resistance of the current study may be attributed to different parameters, such as the adhesive system or the photopolymerization shrinkage stresses. In the LV group, a self-adhesive resin cement (RelyX Unicem) was used for the adhesion of the composite inlays to tooth tissues, while in the BF group, an etch-and-rinse adhesive system (Single Bond Universal) was used. Both adhesive techniques were previously investigated and showed sufficient bond strength to dental tissues.39,40
Regarding the fracture mode distribution between the groups, the statistical analysis did not reveal any significant differences. The most common types of failures were found to be types III and IV, which were related to the tooth restoration interface as the fractures appeared in the buccal and lingual cavity walls and the axial-pulpal angle. These types could be repaired intraorally since the fracture line was located above the cemento-enamel junction (13 specimens of the BF group and 13 specimens of the LV group), while only seven (four specimens of the BF group and three specimens of the LV group) were considered impossible to be repaired. Factors that may affect the type of fracture could be the adhesive procedures, the fixation of the casts, the direction and angle of the force exerted, and the type of selected antagonist.41
Regardless of the fracture type, there are controversial views on the cusp protection in relation to the range of the occlusal isthmus. There are studies recommending cusp protection in order to have a sustainable restoration when the thickness of the remaining cavity walls reaches 2 mm or less.24,42 On the other hand, Fonseca and others26 reported that the cavity preparation and more particularly the width of the occlusal isthmus did not affect the fracture resistance of teeth bearing prepolymerized composite resin restorations, while Silva and others20 claimed that extended dental preparations can be restored with prepolymerized composite resin restorations even in patients with a strong chewing activity. The results of this study indicated that cavities with a wall thickness of approximately 2 mm can be restored with RC CAD/CAM inlays, demonstrating a fracture strength that is higher than the stresses found in the oral cavity.
Within the limitations of this study, the first null hypothesis was rejected, as the results indicated that there was a significant difference of the fracture resistance between the experimental groups. The indirect CAD/CAM technique (LV group) presented significantly higher fracture resistance compared to that of the direct bulk-fill technique (BF group). The results are in agreement with previous studies comparing conventional direct composite restorations with indirect CAD/CAM composite restorations.20,24 Nevertheless, the comparison between a bulk-fill and a CAD/CAM composite restoration has never been investigated before. As expected, intact teeth exhibited higher fracture resistance than the restored teeth, which is in accordance with previous studies evaluating various types of cavity preparations.25,26 The second null hypothesis of the study was accepted, as it was found that there was no significant difference in the mode of fracture between the two experimental groups.
Although fatigue tests provide more accurate data for restored teeth failures, they present difficulties in repeatability and are more time consuming.20,43 Fracture resistance tests, despite their limitations, provide valuable information on the ability of restored teeth to cope with specific clinical conditions.26 One of the main disadvantages of fracture resistance tests compared to fatigue tests is that the fracture load of a specimen, which breaks under aggravating conditions, offers limited information on preexisting tooth cracks.44 Therefore, before engaging any time-consuming durability tests, it was necessary first to establish that the structural integrity of the restored tooth had not been fatally compromised by the use of an untested design. This study has shown that this is not the case and that teeth restored with both restorative materials fail after application of a load that is a lot higher than the loads exerted in the oral cavity. As an indication, this is promising but requires further investigation with fatigue tests before suggesting the restoration of cavities with extensive occlusal isthmus with no cusp protection and with materials and techniques tested in this study as an option for the clinician.
CONCLUSIONS
Within the limitations of this in vitro study, it can be concluded that CAD/CAM RC inlays could be promising alternative restorative materials for improving the prognosis of large posterior tooth restorations with regard to their fracture resistance. Although the bulk-fill composite restorations presented lower fracture resistance when compared to CAD/CAM composite materials, they can be considered appropriate for use in large posterior restorations, as the restorations failed at loads a lot higher to those found in the oral cavity. The mode of fracture showed that the majority of failures reported for both direct and indirect restorations could be repaired intraorally. However, further clinical studies are necessary to confirm the significance of the results of the present investigation. Fatigue tests should also be useful to evaluate the mechanical properties of these types of restorations.
Representative mesio-occlusal-distal cavity preparation ready for being restored.
Layout of the tooth specimen and stainless-steel device used for fracture resistance testing.
Boxplot method, which reveals the outlier data. Figure 4Means and standard deviations of fracture resistance (N) of the experimental groups of the study.
(a-e): Representative images of the five types of fracture mode of restored teeth according to Burke's classification. (a): Type I. (b): Type II. (c): Type III. (d): Type IV. (e): Type V.
Contributor Notes