Tooth Selection and Cavity Preparation

Forty-five extracted, intact, caries-free human third molars were used (Ethics Committee in Human Research approval no. 06257012.1.0000.5152). The teeth were selected to have an intercuspal width within a maximum deviation of 10% from the determined mean.¹⁴ The intercuspal width varied between 4.81 mm and 5.98 mm. The roots of the teeth were covered with a 0.3-mm layer of a polyether impression material (Impregum, 3M ESPE, St Paul, MN, USA) to simulate the periodontal ligament and then embedded in a polystyrene resin (Cristal, Piracicaba, Brazil) up to 2 mm below the cementum-enamel junction to simulate the alveolar bone.¹⁵

The teeth were cleaned using a rubber cup and fine pumice water slurry. Class II mesio-occlusal-distal (MOD) cavities with 4/5 of the intercuspal width and 5-mm depth were prepared in all specimens with a diamond bur (#3099 diamond bur, KG Sorensen, Barueri, Brazil) with copious air-water spray using a cavity preparation machine.¹⁶ This machine consisted of a high-speed handpiece (Extra torque 605 C, KaVo do Brasil, Joinvile, Brazil) coupled to a mobile base. The mobile base moves vertically and horizontally with three precision micrometric heads (152-389, Mitutoyo Sul Americana Ltda, Suzano, Brazil), attaining a 0.002-mm level of accuracy. The endodontic access was manually performed with a diamond bur (#1016 HL KG Sorensen) in a high-speed handpiece with copious air-water spray.

The composition and manufacturer information of the restorative materials used in this study are listed in Table 1. The specimens were distributed into three groups (n=15). In the composite resin (CR) group teeth were restored with an incremental filling technique using composite resin (Filtek Supreme, Shade A2, 3M-ESPE, St Paul, MN, USA) to fill the pulp chamber and MOD cavity. In the resin modified glass ionomer cement in combination with composite resin (RMGI-CR) group, teeth were restored using resin modified glass ionomer cement to fill the pulp chamber (Vitremer, Shade A2, 3M-ESPE) followed by the same composite resin technique to fill the MOD cavity; In the conventional glass ionomer cement in combination with composite resin (CGI-CR) group, teeth were restored using conventional glass ionomer cement to fill the pulp chamber (Ketac Fil, 3M-ESPE) followed by the same composite resin technique to fill the MOD cavity. Ten restored teeth per group were used for measuring cuspal deflection with strain gauges and afterward for compressive fracture resistance test. The other five restored teeth were used for Vickers hardness (VH) and elastic modulus (EM) measurements using the dynamic indentation method.

Table 1:  Materials Tested in the Study ^a

View Fullscreen

Cuspal Deformation

Cuspal deformation was measured with strain gauges (PA-06-060CC-350L, Excel Sensores, Embú, Brazil), which had an internal electrical resistance of 350 Ω, a gauge factor of 2.07, and a grid size of 21.02 mm². The gauge factor is a proportional constant between electrical resistance variation and strain. The strain gauges were placed on 10 teeth in the region where a finite element model had indicated the presence of the highest polymerization strains.¹⁷ One strain gauge was placed on the external surface of the lingual cusp next to the height of the pulp chamber. The other strain gauge was placed on the external surface of the buccal cusp next to the base of the Class II MOD cavity. In addition, two strain gauges were fixed to another intact tooth to compensate for dimensional deviations due to temperature effects. The strain gauges were bonded with cyanoacrylate-based adhesive (Super Bonder, Loctite, Itapeví, Brazil), and the wires were connected to a data acquisition device (ADS0500IP, Lynx, São Paulo, Brazil).

For the RMGI-CR and CGI-CR groups, the glass ionomer cements were manipulated as recommended by the manufacturer and used to fill the pulp chamber, thereby creating a flat surface of the pulpal floor level with the gingival wall of the proximal boxes. The conventional glass ionomer was covered with wax for 5 minutes to achieve complete setting. The resin modified glass ionomer was light cured for 40 seconds using a light source with 600 mW/cm² output (Demetron Kerr, Orange, CA, USA). Selective etching of enamel was done for 15 seconds, and a two-step self-etching adhesive system (Clearfil SE Bond, Kuraray America, New York City, USA) were used for hybridization procedures in all groups. The average volume of composite per increment used was 24.3 mm³ (no more than 2-mm thick) defined by a Teflon matrix to standardize before insertion into the cavity. For the CR group, the pulp chamber was filled in two resin increments and the coronal cavity for all groups was filled with eight resin increments (two at the mesial box, two at the distal box, and four at the occlusal box). Each increment was light cured for 40 seconds from the occlusal direction closest to the cavity.

The cuspal deformation data were obtained from the strain gauges through data analysis software (AqDados 7.02 and AqAnalisys, Lynx Tecnologia Eletrônica, São Paulo, SP, Brazil). The strain values were recorded at 4 Hz during the restorative procedure and continued for 10 minutes after curing the last increment.

Fracture Resistance Test

After cuspal deformation measurements, teeth were subjected to axial compressive loading with a metal sphere 6 mm in diameter at a crosshead speed of 0.5 mm/min in a universal testing machine (DL2000, EMIC, São Jose dos Pinhais, Brazil). The load required (N) to cause catastrophic fracture of specimens was recorded by a 500 N load cell hardwired to a computer with control and data acquisition software (TESC 3.04, EMIC). The failure mode of each specimen was analyzed under a stereomicroscope (Leika Ecafix, Tokyo, Japan) at 40× magnification, and then assigned to one of four categories, using a modified classification system based on that proposed by Burke¹⁸: (I) fractures involving a small portion of the coronal tooth structure; (II) fractures involving a small portion of the coronal tooth structure and cohesive failure of the restoration; (III) fractures involving the tooth structure, cohesive and/or adhesive failure of the restoration, with root involvement that can be restored in association with periodontal surgery; and (IV) severe root and crown fracture, which determine extraction of the tooth.

Vickers Hardness and Elastic Modulus

The remaining five specimens from each group were used for analysis of the mechanical properties (VH and EM) of the composite resin/glass ionomer restorations at eight depths. Each restored tooth was sectioned in the buccal-lingual direction into two halves using a precision saw (Isomet 1000, Buehler, Lake Bluff, IL, USA). One section per tooth was randomly selected for assessment of the mechanical properties. The specimens were embedded with methacrylate resin (Instrumental Instrumentos de Medição Ltda, São Paulo, Brazil). Before testing, the surfaces were finished with silicon-carbide papers (#600, 800, 1200, and 2000 grit, Norton, Campinas, Brazil) and polished with metallographic diamond pastes (6-, 3-, 1-, and 1/4-μm, Arotec, São Paulo, Brazil). After polishing, the specimens were cleaned using ultrasound for 10 minutes with distilled water. Using a Vickers indenter (CSM Micro-Hardness Tester, CSM Instruments, Peseux, Switzerland), indentations were made every 1.0 mm from 0.5 mm to 7.5 mm, starting from the base of the pulp chamber of the restorations. The indentations were carried out with controlled force, whereby the test load was increased or decreased at a constant speed ranging between 0 and 500 mN in 20-second intervals. The maximum force of 500 mN was held for 5 seconds. The load and the penetration depth of the indenter were continuously measured during the load-unload hysteresis. The universal hardness was defined as the applied force divided by the apparent area of the indentation at the maximum force. The measurements were expressed in VH units by applying the conversion factor supplied by the manufacturer. The indentation modulus was calculated from the slope of the tangent of the indentation depth curve at the maximum force and is comparable to the material's EM.¹⁹

Post-Gel Shrinkage

Post-gel linear shrinkage was determined using the strain gauge method.²⁰ Ten specimens were tested for each restorative material. The materials were shaped into a hemisphere on top of a biaxial strain gauge (CEA-06-032WT-120, Measurements Group, Raleigh, NC, USA) that measured shrinkage strains in two perpendicular directions (X and Y axes). A strain conditioner (ADS0500IP, Lynx Tecnologia Eletrônica) converted electrical resistance changes in the strain gauge to voltage changes through a quarter-bridge circuit with an internal reference resistance. The strains measured along the two axes were averaged because the material properties were homogeneous and isotropic on a macro scale. Filtek Supreme (composite resin) and Vitremer (RMGI) were light-cured using a quartz-tungsten-halogen unit (Demetron, Kerr, Orange, CA, USA) with the light tip held at a 1-mm distance from the surface of the composite and monitored for 10 minutes. The radiant exposure was set at 18 J/cm² (600 mW/cm² ×30 s). Ketac Fil (conventional glass ionomer) was positioned on the strain gauge after mixing and was monitored for 30 minutes. The mean shrinkage strain was used as linear post-gel shrinkage input for the finite element analysis and could be converted to volumetric percentage by multiplying by 3 and 100%.

Residual Stress Calculation: Finite Element Analysis

To calculate corresponding residual stresses in the tooth, a two-dimensional (2D) finite element model was analyzed for the MOD restoration with endodontic access. The geometry of the model was based on a digitized buccolingual cross-section of a third molar with similar dimensions as the teeth selected for laboratory tests. Coordinates were obtained using digital image processing software (ImageJ 1.48, National Institutes of Health, Bethesda, MD, USA). The applied mechanical properties were EM of enamel 84 GPa and Poisson ratio 0.30 as well as EM of dentin 18 GPa and Poisson ratio 0.23.²¹ The EM values of the three restorative materials measured in this study were also used for the analysis (Table 4). The Poisson ratios used were 0.24,²² 0.35,²³ and 0.33,²⁴ respectively for Filtek Supreme, Vitremer, and Ketac Fil. A plane strain condition was assumed for the tooth cross-sections. Because of this 2D strain condition and consequently 2D finite element model, no distinction was made between the mesial and distal increments.

Table 4:  Elastic Modulus (GPa) and Vickers Hardness (N/mm²) Averaged from Eight Measurement Points^ba

View Fullscreen

The finite element analysis (FEA) was performed using MSC.Mentat (preprocessor and postprocessor) and MSC.Marc (solver) software (Version 2010r2, MSC Software Corporation, Santa Ana, CA, USA). One FEA model was generated for each experimental group to simulate the three restorative protocols of the experimental study. Polymerization shrinkage was simulated by thermal analogy. Temperature was reduced by 1°C, while the linear shrinkage value (post-gel shrinkage) was entered as the coefficient of linear thermal expansion. Modified von Mises equivalent stress was used to express the stress conditions using compressive-tensile strength ratios of 37.3,²² 3.0,²² and 5.82,¹⁴ respectively for the enamel, dentin, and Filtek Supreme as well as 3.55,²⁵ and 11.3,²⁶ respectively, for the Vitremer and Ketac Fil. Stress values were recorded in the integration points of each element and in nodes along material interfaces at either aspect (tooth and restoration).

Statistical Analysis

The cuspal deformation (strain), fracture resistance, post-gel shrinkage, EM, and VH data were tested for normal distribution (Shapiro-Wilk, p>0.05) and equality of variances (Levene test), followed by parametric statistical tests. One-way analysis of variance (ANOVA) was performed for cuspal deflection in each cusp, fracture resistance, and post-gel shrinkage. One-way ANOVA was performed in a split-plot arrangement, with the plot represented by restorative protocol and the subplot represented by the depth of the cavity. Multiple comparisons were made using a Tukey test. The data of fracture mode were subjected to a χ² test. All tests used α = 0.05 as the significance level, and all analyses were carried out with the statistical package Sigma Plot version 13.1 (Systat Software Inc, San Jose, CA, USA).

Cuspal Strain (CS)

The values of cuspal deformation (strain) for the three filling techniques are shown in Figure 1 and Table 2. For the lingual cusp, the CR filling technique had the highest cuspal deformation values, followed by the RMGI-CR and CGI-CR techniques, respectively. For the buccal cusp, the CR filling technique also had the highest cuspal deformation values. Buccal deformation was not significantly different between the RMGI-CR and CGI-CR groups.

View Figure

• Download as Powerpoint
• Download Figure

Table 2:  Cuspal Deformation (μS) Measured by Strain Gauges and Fracture Resistance ^a Measured by Axial Compression Test (n=10)

View Fullscreen

Fracture Resistance and Failure Mode

The mean fracture resistance and standard deviation for the three restorative techniques are shown in Table 2. One-way ANOVA showed significant difference among groups (p=0.02). The CR group had significantly lower fracture resistance than the other two filling techniques that used glass ionomer to fill the pulp chamber. No differences were found between the RMGI-CR and CGI-CR groups. Failure mode distribution is shown in Table 3. No significant differences in failure modes were found among the three groups.

Table 3:  Failure Mode Distribution ^a

View Fullscreen

Vickers Hardness

The VH of the three filling techniques at various depths of the restorations are shown in Figure 2A and Table 4. The VH of the CR filling technique was constant through the entire restoration. For the RMGI-CR and CGI-CR filling techniques, the VH was constant on coronal restoration (composite resin portion) and decreased significantly in the pulp chamber (glass ionomer–based materials). No differences were found between the VH on the coronal portion of the three filling techniques (Table 4). The VH measured in the pulp chamber was significantly lower for the RMGI-CR and CGI-CR restorative techniques (Table 4).

View Figure

• Download as Powerpoint
• Download Figure

Elastic Modulus

The EM of the three filling techniques at various depths of the restorations are shown in Figure 2B and Table 4. The EM of the CR filling technique decreased significantly only after the 6.5-mm depth. For the CGI-CR and RMGI-CR filling techniques, the EM was constant in the coronal restoration (composite resin portion). The EM decreased significantly in the pulp chamber for the RMGI-CR and CGIRC techniques (Figure 2B). The EM values measured in the pulp chamber were significantly lower for the RMGI-CR and CGI-CR groups (Table 4).

Post-Gel Shrinkage

The mean values and standard deviations for the post-gel shrinkage of three restorative materials are presented in Table 5. One-way ANOVA revealed statistical difference among the composites (p<0.001). Filtek Z350 XT had significantly higher mean volumetric shrinkage value than Ketac Fil and Vitremer.

Table 5:  Mean (± Standard Deviation) Volumetric Post-Gel Shrinkage ^a

View Fullscreen

Finite Element Analysis

Stress distribution for all groups is shown in Figure 3. The filling technique influenced the stress distribution and stress intensity. The CR filling technique resulted in higher stresses in the enamel and in root dentin close to the pulp chamber than the CGI-CR and RMGI-CR filling techniques. For comparison, Figure 4 shows cuspal deformation calculated with the FEA in the same buccal and lingual locations where the strain gauges were placed in the experiments (Figure 1). Experimentally measured cuspal deformation values were similar to the deformation values calculated by FEA.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Microhardness

Hardness has been used as an indicator of degree of conversion (ie, extent of polymerization of monomers to polymers) and curing depth in resin-based dental materials.^31-33 Filtek Supreme inserted incrementally showed constant values of VH at all depths of the restorations, suggesting adequate polymerization and degree of conversion.¹⁴ The Ketac Fil and Vitremer had similar VH, which was significantly lower than that of Filtek Supreme. This may be explained by the different compositions of these materials. Glass ionomer cements are mechanically weaker compared with composite resins because of the weak bonding between the particles of glass and polyacid matrix.³⁴ Resin-modified glass ionomers essentially consist of conventional glass ionomer components combined with organic photopolymerizable monomers for the initial setting reaction.³⁵ The VH values of the materials in the pulp chamber were constant for both glass ionomers, demonstrating that their curing process was efficient. The VH of the Filtek Supreme is very similar to the value for human dentin. However, when a correlation is made with the intact tooth, the pulp chamber is similar to an empty, not a load supporting, space. The use of materials with lower stiffness may thus explain the better biomechanical performance of the two groups that used glass ionomer to fill pulp chamber.

Elastic Modulus

The EM represents the relative stiffness of a material and gives information about how occlusal forces will be supported by the materials. The incrementally inserted Filtek Supreme showed constant values of EM up to 6.5 mm depth. At the bottom of the restoration the VH was maintained but EM decreased, demonstrating that VH and EM have no direct correlation.¹⁴ The distance between the light positioned, even with limited access by the cusps was apparently sufficient to cure the very deep region of the cavity but not to attain the same EM. Filtek Supreme had the same EM in the coronal restoration in all tested groups, confirming that the biomechanical differences found among the three groups were related to the restorative material in the pulp chamber. The low EM of glass ionomers can be explained by their composition. Particle type and size as well as the flexible matrix of glass ionomers contribute to their relative weakness.^34,36,37

Cuspal Strains

Cuspal deformation in restored teeth is affected by many factors, including the size of the cavity, the properties of the restorative material, and the filling technique.¹⁴ In this study the size of the cavity was standardized for all specimens. Assuming similar tooth properties, shapes, and sizes, the main variables causing differences in cuspal deformation were therefore the properties of the materials and the filling techniques. We used strain gauges to measure cuspal deformation. Because strains vary across the tooth surfaces, the attachment location is important for the results. FEA was used to identify the areas where the highest strains occurred. The lingual cusp had more cusp deformation than the buccal cusp, irrespective of the restorative protocol. This result can be explained by the amount of remaining tooth structure. Third molars had thinner cervical areas lingually than they had buccally, and, thus, the lingual cusps can be expected to be less stiff and deform more than the buccal cusps. The cuspal deformation values were significantly different among the restorative protocols tested. Using a glass ionomer material to fill the pulp chamber and overlaying it with composite resulted in lower cuspal strain than using composite for the whole restoration. The reduction this gave in cuspal strains can be attributed to the lower shrinkage in combination with the lower EM of the glass ionomers. For the same reason glass ionomers have been recommended as cavity lining to reduce shrinkage stress of the restoration.²⁸

Fracture Resistance

The fracture resistance values showed significant differences between the restorative protocols, with significantly lower resistance for the CR group. According to Reeh and others,²⁷ endodontic procedures have only a small effect on the tooth, reducing the relative rigidity by 5%, which is contributed entirely to the access opening. Restorative procedures and, particularly, the loss of marginal ridge integrity were the greatest contributors to the loss of tooth rigidity. Despite the lower mechanical properties of glass ionomer cement compared with composite resin, placement of a glass ionomer into the pulp chamber resulted in higher fracture strength than restoration with composite alone. A relatively stiffer restoration, in this case with the all-composite resin technique (CR), will transfer and distribute the occlusal loads deeper into the restoration and pulp chamber,³⁸ making the endodontically treated tooth more vulnerable around the compromised pulpal area. Furthermore, anatomically the lingual cusp seems to be more prone to fracture, which has been attributed to an unfavorable distribution of stresses during mastication.³⁹ The failure modes observed in this study confirm that most of fractures were concentrated on the lingual cusp (failure mode III). This area was also where shrinkage stresses were more concentrated along the pulp chamber, which may have contributed to this type of fracture.

Validation

The validation and correlation of experimental and computational methods is an important step in a comprehensive research approach and is essential to justify conclusions drawn from in vitro analyses.⁶ Although FEA was essential to assess the stress conditions, the validity of stress calculations depends on the correct input of material properties, anatomic shape, and restraints of the restored tooth structure.⁴⁰ A validated finite element model can be further used to predict mechanical failures or investigate questions that cannot be accessed as well in laboratory tests.⁶ Although the calculated stresses could not be validated directly from the laboratory experiments, they could be verified indirectly from the deformation and its consequences.⁴⁰ In our study, cuspal strains calculated by the FEA were similar to the cuspal strain data collected experimentally using strain gauges placed on cuspal surfaces (Figures 1 and 4). This close similarity supports the validity of our FEA models and stress results.^38,40,41

Shrinkage stress is a serious concern, as has been demonstrated clinically by a high incidence of secondary caries in endodontically treated teeth restored with resin composites over a 5-year period.¹² The finding of this study could be important when direct composite restoration is the treatment of choice because of cost or conservative approach. Using conventional or resin modified glass ionomer to fill the pulp chamber under composite restoration may improve the longevity of the endodontically treated molars.