INTRODUCTION
Light curing composite materials are widely used in clinical dentistry as restorative materials due to their esthetic, mechanical and handling properties. If a composite material is not polymerized properly, it can lead to the leaching of components either from filler or mostly, from unpolymerized organic matrix. Even properly cured composite materials contain a certain amount of residual monomers that can be eluted and exert a toxic effect (1). Factors like filler composition, filler content, filler surface area and type of filler particle treatment process can influence the amount of leached monomers (1, 2). Ferracane and Condon (3) showed in their study that the highest cytotoxicity induced by unpolymerized composites occurs during the first 24 h. However, Wattha et al., (4) tested cytotoxicity of resin-containing restorative materials after aging in artificial saliva and concluded that resin-based restorative materials may release residual components, which trigger cytotoxicity for up to 2 weeks. In vitro studies indicated that methacrylate and dimethacrylate monomers used in restorative dental materials may increase the intracellular level of reactive oxygen species which induce apoptosis. In addition, they suppress the mitochondrial activity of macrophages, thus altering their inflammatory responses, affect the recruitment of leukocytes and decrease the expression of intercellular adhesion molecules, induce enzymatic activity, DNA fragmentation, expression of growth factors and cytokines (5-9).
Theoretically, a 100% conversion of monomers to polymers is possible, but usually 25-50% of methacrylate monomer double-bonds remain unreacted and it is estimated that 5-10% of the total amount of unreacted C=C bonds are available for interaction with macromolecules in a biological system (3, 10, 11). Composite materials do not reach the maximum degree of conversion immediately after light curing (12, 13). Moon et al., (5) reported that a period of 7 days is needed for the degree of conversion of the materials to reach maximum polymerization on both bottom and top surfaces. Further, they concluded that the amount of monomers leached from the same composite material differed in regards to the type of the curing unit and the curing method. In another study, Moon (14) shows that differences between the amount of leached monomers and mechanical properties occurred when the radiant energy emitted from the curing units is lower than 17 J/cm2. In contrast, if the radiant energy is higher than 17 J/cm2, those differences disappear regardless of the irradiation programs and curing units used (15).
Recently, the pre-heating of composite materials before their application in the oral cavity became an acceptable approach in order to obtain a higher degree of conversion and better mechanical properties without negative effect on marginal seal (16, 17). The use of pre-heated composite as a luting material for CAD/CAM restorations was also reported (18-20).
Daronch et al. (17) reported that pre-heating the composite up to 60 oC may improve the degree of conversion on both, top and 2 mm deep surface. However, Froes-Salgado et al., (21) did not find any improvement in mechanical properties and the degree of conversion of pre-heated composite, but reported an improvement in composite adaptation to cavity walls. Lohbauer et al. (22) indicated that pre-heating of composite materials may have a negative effect on the restoration margins because of the higher polymerization shrinkage due to a higher degree of conversion.
Since the use of pre-heated composite materials as a luting material for CAD/CAM fabricated restorations are becoming more and more popular, it will be interesting to see the impact of the light curing of heated composite through CAD/CAM restoration on cytotoxiticy/ genotoxicity. Therefore, the aim of this study was to assess cytotoxicity and genotoxicity of two different composite resins pre-heated at three different temperatures and light-cured directly and through 2 mm thick CAD/CAM onlays. For that purpose, the following null-hypotheses are formed:
MATERIALS AND METHODS
Specimen Preparation
A two mm thick layer of a ceramic-reinforced polymer (CRP, LAVA Ultimate, 3M ESPE, St. Paul, MN, USA) CAD/CAM of shade A2 (block size 14; 14x14x18 mm) and lithium disilicate glass-ceramic (LDC, e.max CAD, Ivoclar Vivadent, Schaan, Liechtenstein) of shade A2 (block size 14; 14x12x17 mm), were used as overlays for light-curing of composite samples. CRP and LDC blocks were sectioned using a water-cooled precision diamond saw. The overlay sample size was 14x14x2 mm for CRP and 14x12x2 mm for LDC. Thereafter, the samples were metallurgically polished to high gloss (#600, #800, #1200 grit). For LDC samples, each sample was glazed (Crystall/Glaze Spray; Ivoclar/Vivadent, Schaan, Liechtenstein) according to the manufacturer instructions.
Two different composite resins were used: a microhybrid composite (Z100, 3M ESPE, St. Paul, MN, USA) and a nanofilled composite resin (Filtek Supreme Ultra, 3M ESPE). The composition of these materials used in this study is given in Table 1. The composite resins were pre-heated using a heating unit (Calset, AdDent Inc., Danbury, CT, USA) at three different temperatures: 37 oC (T1), 54 oC (T2), and 68 oC (T3) according to the manufacturer instructions. A tray placed on the top of the heater contains slots to place the composite compules. In approximately 10 minutes the desired temperature is reached and the composite is ready to use.
The samples for cyto-/genotoxicity testing were prepared as follows: A mold of 6 mm in diameter and 0.65 mm in thickness was positioned on a round stainless steel disc (diameter 6 mm, thickness 5 mm), which was covered with a clear Mylar matrix. The mold was then carefully filled with the uncured composite material ensuring that no air bubbles were left. The composite specimens were covered with another Mylar sheet and pressed with another round stainless steel disc (diameter 6 mm, thickness 5 mm) to obtain a homogenous thickness of the composite sample (0.65 mm). The stainless steel disc was removed and Mylar matrix remained on the sample to prevent formation of the oxygen-inhibited layer on the surface of polymerized composite material. All tested composite resin samples were polymerized with a light-emitting diode (LED) light curing unit (Bluephase G2, Vivadent, Schaan, Liechtenstein) using high intensity mode (1180 mW/cm2) for 40 sec. Three light-curing modes were used: (1) direct curing; (2) curing through CAD/CAM CRP overlay, and (3) curing through LDC overlay. Once the light-curing was completed, the specimens were removed from the mold and introduced into cell cultures.
For the preparation of unpolymerized sample, 0.06 g of the composite material was taken and introduced directly into the cell culture.
Primary lymphocyte cultures
This study was approved by the Ethical Committee, School of Dental Medicine, University of Zagreb, Croatia. Primary lymphocyte cultures were set from the cells of a single donor, 39 year old male, non-smoker, with no medical records of chronic or acute adverse health conditions. To overcome possible inter-individual differences in response to the treatment we applied a single donor sampling approach. Prior to blood sampling, the donor was acquainted with the procedure, purpose of blood donation, and aims of the testing the blood is to be used for.
Blood was collected by antecubital venipucture in heparinized vacutainers (Becton Dickinson, UK). Lymphocytes were isolated as described previously by Kopjar et al. (23). Following the isolation, 50,000 lymphocytes were seeded in sterile tubes (Nange Nunc Int, Naperville, IL, USA) in RPMI 1640 culture medium (Gibco Invitrogen, UK), with a final culture volume to be 7 ml.
Each tested material (0.06 g), both in unpolymerized and polymerized form, was placed in the lymphocyte culture. Treatments lasted for 24 h at 37 oC in 5% CO2 atmosphere (Heraeus Hera Cell 240 incubator, Langenselbold, Germany). Following the 24 h of treatment, cultures were centrifuged at 300 g, 5 min. Supernatant was removed and pellet containing the lymphocytes was resuspended and used for further analyses.
Quantitative fluorescent assay for the assessment of cell viability, apoptosis and necrosis
The viability of peripheral blood lymphocytes was assessed using a dye exclusion method (24). In this assay, viable (intact plasma membrane) and dead (damaged plasma membrane) cells can be visualized after simultaneous staining with the fluorescent DNA-binding dyes ethidium bromide and acridine orange.
A mixture of ethidium bromide and acridine orange (Sigma-Aldrich, USA) in final concentrations of 100 µg/ml (1:1; v/v) was gently pippeted onto the lymphocyte suspension (V=20 µL) placed on a microscope slide, covered with a coverslip and immediately analyzed under a fluorescence microscope (Olympus BX; 400 x magnification). Three tests with aliquots of the same sample were performed and a total of 300 cells per sample were counted. Control, untreated lymphocyte culture was studied in parallel. Quantitative assessments were made by the determination of the percentage of viable, apoptotic and necrotic cells. Viable cells excluded ethidium bromide and the appearance of their nuclei with an intact structure was bright green. Non-viable necrotic cells had orange to red colored chromatin with organized structure while apoptotic cells were bright green with highly condensed or fragmented nuclei.
Alkaline comet assay
Ten µl of lymphocyte resuspension was mixed with 100 µl of 5% low melting point agarose (37 oC; Sigma–Aldrich, MO, USA) and placed onto normal melting point (Sigma–Aldrich, MO, USA) precoated microscope slides, covered with a slip cover, and let to polymerize. Slides were immersed into lysis buffer (2.5 M NaCl, 0.1 M Na2EDTA, 10 mM Tris–HCl, 1% N-lauroylsarcosine, 10% DMSO, 1% Triton X-100; Sigma-Aldrich, MO, USA; pH 10) for 20 min at 4 oC. Slides were denaturated in buffer (1 mM Na2EDTA, 300 mM NaOH; Sigma-Aldrich, MO, USA; pH > 13 for 20 min) and subjected to electrophoresis using a buffer of the same composition as the one used for denaturation. Electrophoresis lasted for 20 min at 0.7 V/cm. Slides were analysed under epifluorescent microscope Olympus BX 51 (Olympus, Japan) connected to Comet Assay IV analysis system (Perceptive Instruments, UK). A total of 50 comets per treatment were scored in duplicate. Results are expressed using tail length (µm) and tail intensity (% of DNA in comet tail) and presented as mean and median and S.D. of two scorings.
Prior to immersion in lysis buffer, as the positive control, slides obtained from untreated lymphocyte cultures were treated with 60 µl of 1 mM H2O2 for 10 min placed on ice.
Statistical analysis
Comparisons between the values observed for cell viability, apoptosis, and necrosis were performed by Pearson’s χ2-test for two-by-two contingency tables. The data acquired by alkaline comet assay were first evaluated using descriptive statistics (Microsoft Excel). More detailed statistical analysis was performed with the statistical software (Statistica 10, StatSoft, OK, USA). The data were first transformed by applying log transformation to normalize the distribution (25). A one-way analysis of variance (ANOVA) was computed, followed by a Tukey post hoc test (α < 0.05).
RESULTS
Microhybrid Composite: Z100
Results of the quantitative fluorescent assay for simultaneous identification of apoptotic and necrotic cells in lymphocyte samples incubated with Z100 are reported in Table 2. After 24 h of incubation with unpolymerized Z100 there were 88.7±2.1% viable lymphocytes, while in the negative control lymphocyte viability was 97.7±0.6% (P<0.0001). Cytotoxicity of the Z100 subjected to T1sec preheating was generally lower when compared with that of the unpolymerized Z100, irrespective of the light-curing mode (direct, through CRP or LDC). Statistically significant differences were found between unpolymerized and directly light-cured Z100 (P=0.0309). The best results with the highest percentages of viable cells were observed for T1 preheating: 93.7±0.6% of viable cells after direct light-curing; 92.3±2.1% of viable cells after light-curingthrough CRP, and 91.7±0.6% of viable cells after light-curing through LDC CAD/CAM overlay (Table 2). In the negative control and unpolymerized Z100 the frequency of apoptotic lymphocytes was slightly higher than the frequency of necrotic lymphocytes. Apoptosis predominated over necrosis only in samples of Z100 prepared with direct light-curing regardless of the temperature applied for preheating. Increase of preheating temperature did not significantly influence the frequency of necrosis in tested samples. The only significant differences were found at T1, where in directly light-cured sample the lowest percentage of necrotic cells was found.
Note.
300 cells per sample per each experimental point were analysed.
Statistical significance of data was evaluated using χ2 test. Significant differences (P<0.05) are indicated in the table; NC – vs. negative control; UN – vs. unpolymerized material; T1 – vs. sample exposed to the same polymerized material, preheated at T1; T2 – vs. sample exposed to the same polymerized material, preheated at T2; T3 – vs. sample exposed to the same polymerized material, preheated at T3.
Nanofilled Composite: FILTEK SUPREME ULTRA
Results of the quantitative fluorescent assay for the simultaneous identification of apoptotic and necrotic cells in lymphocyte samples incubated with Filtek Supreme Ultra are reported in Table 3. After 24 hours of incubation with unpolymerized Filtek Supreme Ultra there were 89.7±2.1% viable lymphocytes, while in the negative control lymphocyte viability was 97.7±0.6% (P<0.0001). Cytotoxicity of the polymerized material, regardless of the light-curing mode (direct, through 2 mm thick CRP CAD/CAM overlay and through 2 mm thick LDC CAD/CAM overlay) and temperatures of preheating, in all cases was significantly higher compared with the negative control (P<0.01). Although the minor differences in the percentages of viable cells between samples preheated at three temperatures using different light-curing modes were observed, none of them was statistically significant compared to unpolymerized material.
{ label needed for table-wrap[@id='t3'] }
Note.
300 cells per sample per each experimental point were analysed.
Statistical significance of data was evaluated using χ2 test. Significant differences (P<0.05) are indicated in the table; NC – vs. negative control; UN – vs. unpolymerized material; T1 – vs. sample exposed to the same polymerized material, preheated at T1; T2 – vs. sample exposed to the same polymerized material, preheated at T2; T3 – vs. sample exposed to the same polymerized material, preheated at T3.
Preheating at T1 resulted with the highest percentage of viable cells in sample which was overlayed with 2 mm thick LDC CAD/CAM block: 90.3±1.5%. Slightly lower viability was observed after light-curingthrough 2 mm thick CRP CAD/CAM overlay (86.7±1.5%), while the lowest lymphocyte viability (84.7±3.2%) was observed after direct light-curing. In the negative control and unpolymerized Filtek Supreme Ultra, the frequency of apoptotic lymphocytes was slightly higher than the frequency of necrotic lymphocytes. Apoptosis predominated over necrosis in almost all polymerized samples, regardless of the temperature applied for preheating. The lowest frequency of apoptosis was observed after preheating at T2; with the best result in sample light-cured through 2 mm thick LDC CAD/CAM overlay (4.7±0.6% of apoptotic cells). An increase of preheating the temperature in a majority of the tested samples did not significantly influence the frequency of necrosis, as compared to unpolymerized material. However, in almost all samples, the percentages of necrotic cells were significantly higher than in the negative control.
Alkaline comet assay
Results of the alkaline comet assay were used to evaluate primary damage to DNA are presented as a tail length parameter in Table 4 and tail intensity in Table 5. Considering both comet assay parameters, Z100 and Filtek Supreme Ultra in unpolymerized form significantly elevated primary lesions in DNA compared to the control. However, the effect of unpolymerized Z100 is significantly more pronounced than the one of Filtek Supreme Ultra.
NC - statistically significant compared to the negative control Un - statistically significant compared to the results for unpolymerized form of the material of concern Z100 - statistically significant compared to the results for Z100 in the state and polymerization mode of concern Co - statistically significant compared to the results for polymerization through CRP CAD/CAM overlay of the material of concern
Ce - statistically significant compared to the results for polymerization through LDC CAD/CAM overlay of the material of concern
T1 - statistically significant compared to the results for T1 preheating prior to the polymerization of the material of concern
T2 - statistically significant compared to the results for T2 preheating prior to the polymerization of the material of concern
CoTx - statistically significant compared to the results for polymerization through CRP CAD/CAM overlay of the material of concern preheated at indexed temperature
CeTx - statistically significant compared to the results for polymerization through LDC CAD/CAM overlay of the material of concern preheated at indexed temperature
{ label needed for table-wrap[@id='t5'] }
NC - statistically significant compared to the negative control
Un - statistically significant compared to the results for unpolymerized form of the material of concern
Z100 - statistically significant compared to the results for Z100 in the state and polymerization mode of concern
Co - statistically significant compared to the results for polymerization through CRP CAD/CAM overlay of the material of concern
Ce - statistically significant compared to the results for polymerization through LDC CAD/CAM overlay of the material of concern
T1 - statistically significant compared to the results for T1 preheating prior to the polymerization of the material of concern
T2 - statistically significant compared to the results for T2 preheating prior to the polymerization of the material of concern
CoTx - statistically significant compared to the results for polymerization through CRP CAD/CAM overlay of the material of concern preheated at indexed temperature
CeTx - statistically significant compared to the results for polymerization through LDC CAD/CAM overlay of the material of concern preheated at indexed temperature
Considering the indirect polymerization of Z100, tail intensity did not significantly differ in regards to the preheating temperature and polymerization-barrier used (Table 5). However, the tail length was significantly decreased when the material was preheated at 68 oC (T3). Further, light-curing through the CRP CAD/CAM overlay induced slightly lower DNA migration compared to light-curing through LDC CAD/CAM overlay. Statistically significant differences were recorded when temperatures of 54 0C (T2) were used in preheating (Table 4).
For Filtek Supreme Ultra, a significantly lower tail length value compared to other light-curing procedures was obtained by preheating the material at 68 0C (T3) and light-curing it through the LDC CAD/CAM barrier (Table 4). However, the tail intensity parameter did not reveal any statistical significance regarding the preheating temperature or light-curing barrier used (Table 5). Contrary to primary DNA damage measured as tail length, the one measured in terms of % of DNA that migrated into the comet tail did not indicate any difference between the genotoxic potential of two tested composite materials.
In regards to direct light-curing, both materials exhibited a significant genotoxic effect when preheated at 68 oC (T3) prior to light-curing compared to other procedures tested. The effect was recorded as a significant increase in comet assay parameters; both tail length and tail intensity.
DISCUSSION
Since composite restorations are commonly exposed to the oral environment only a few minutes after light-curing, our study mimicked the clinical conditions by incubating cell cultures with composite samples immediately after the sample preparation. The maximum curing time recommended by the manufacturer for the high intensity mode of the LED light curing unit (Bluephase G2) is 15 s, in this study the curing time of 40s was used. The reason for that is when bonding the CAD/CAM restoration, the curing light is attenuated while passing through the restoration and longer exposures have to be done in order to cure the luting composite completely. According to the results of this study, there was no statistically significant difference in cyto-/genotoxicity between light-curing of the composites directly or through 2 mm thick CRP or LDC CAD/CAM overlay. However, the explanation for the results given in this study may be due to the fact that the samples were placed in a lymphocyte cell culture immediately after polymerization. In case of the direct polymerization, the composite sample received more heat than the sample polymerized through 2 mm thick CRP or LDC CAD/CAM sample. Therefore, theheated composite sample when placed in a fresh cell culture may cause less viable cells. This is not because there are unreacted components from the material but because of the temperature of the sample. Our former studies show that the temperature during polymerization plays a crucial role (26-28) but still stays a question since the temperature drops off quickly after removing the composite from the heating unit.
The increase in the level of primary DNA damage was observed for both tested materials. It is unlikely though that the observed effect was mediated by increased temperature of material samples placed in the culture due to pre-heating at 68 0C. Danroch et al. (29) concluded that when composite material is heated up to 60 oC and removed from the heating unit, the temperature of the composite material decreased 35-40% after 40s, up to 50% after 2 min, and up to 90% after 5 min. However, applied heat may affect intramolecular chemical bonds. The radiation absorbed by molecules causes increased vibration. When indirect curing procedures were applied, results obtained for pre-heating temperatures of T2 of 54 oC, which is recommended by composite manufacturers, did not differ compared to the results obtained with pre-heating at T3 - 68 oC.
Some studies reported that the presence of oxygen during the light curing of materials might inhibit polymerization. This leads to the formation of unpolymerized monomeric surface layer which, if not removed after curing, increases cytotoxicity of materials (30, 31). The cytotoxic effect was less exhibited if that surface layer was removed. Therefore, in the present study we used a Mylar sheet to prevent the access of material surface to atmospheric oxygen and to avoid the formation of unpolymerized surface layer and consequently its effect on the cytotoxicity results.
Darmani et al. (32) evaluated cytotoxicity of different composite materials and reported a decrease in cell viabilities up to 48% when the cells were exposed to Z100 composite material. In our study viability was not affected to such extent (Table 2,3{ label needed for table-wrap[@id='t3'] }). Statistical comparison of cytotoxicity results between Z100 and Filtek Supreme Ultra showed that after light-curing of both materials through 2 mm thick CRP CAD/CAM overlay at temperature T1 and T3, better results (higher lymphocyte viability) were obtained for Z100. However, after preheating at temperature T2, Filtek Supreme Ultra had lower cytotoxicity (P=0.0228). Also, after light-curing of both materials through 2 mm thick LDC CAD/CAM overlay at all preheating temperatures, cytotoxicity of Z100 was slightly lower or similar to Filtek Supreme Ultra, however none of the differences was statistically significant. In general, Filtek Supreme Ultra appears less sensitive on how the light-curing is conducted – directly, or through 2 mm thick CRP CAD/CAM sample or 2 mm thick LDC CAD/CAM sample. For both studied materials regardless whether they were cured through CRP CAD/CAM overlay or directly, the lowest cytotoxicity was observed with pre-heating at T1.
Considering both comet assay parameters tail length and tail intensity, the unpolymerized form of Z100 exhibited higher potency of inducing primary damage to DNA compared to Filtek Supreme Ultra. The observed effect is mediated by a significant difference in the composition of two evaluated composite materials. Z100 is TEGDMA and Bis-GMA based material, while Filtek Supreme Ultra besides TEGDMA and Bis-GMA contains also UDMA and Bis-EMA resin monomers. Leaching of residual monomers from Z100 and their toxic effects were reported by several studies (32-34). Darmani et al. (32) concluded that high amount of TEGDMA and comparatively smaller amounts of Bis-GMA were released from tested composite materials. As Engelmann et al. (35) explains, TEGDMA forms adducts with glutathione (which has protective functions to the cells) and may interfere with its protective function leading to the cell destruction. However, Filtek Supreme Ultra contains a considerable amount of zirconia particles, and zirconia is shown to exhibit significant antioxidative properties (1, 36, 37).
In the present evaluation, 24 hours of treatment has been used based on the knowledge gained from previous studies indicating that highest leaching occurs within first 24 hours following restorative material placement in oral cavity (38). Evaluation of the level of primary genomic lesions indicated that CRP CAD/CAM might be a preferable material to perform indirect polymerization, since both tail length and % of DNA in tail were lower compared to LDC polymerization procedure. The percentage of DNA (tail intensity) in the comet tail that is accepted as a more reliable biomarker of genotoxicity directly corresponds to the proportion of genomic DNA affected by the adverse biological effects of the substance (39, 40). The obtained values for that parameter indicates that there is no effect on the primary damage to DNA if preheating is performed at 54 oC (T2) or 68 oC (T3) (Tables 4,5{ label needed for table-wrap[@id='t5'] }). Although there is no statistically significant difference, even considering tail intensity parameter CRP-through polymerization may be preferable procedure (Table 5). The difference in results when polymerized through CPR or LDC CAD/CAM overlay may be explained with different material composition. Ilie and Hickel (41) showed in their study that lithium disilicate glass-ceramic due to is crystalline structure shows more opacity. This can cause less light transmission and therefore less polymerized sample leading to less viable cells.
Regarding the direct polymerization of both tested materials, preheating the procedure at 37 oC (T1) or 54 oC (T2) may be the procedure of choice. Furthermore, in terms of genotoxicity, preheating at 68 oC (T3) may not be recommended.
In general, considering two facts: A) that tail intensity is a more reliable parameter of primary damage to DNA over tail length, and B) that up to 10% of DNA in tail is considered baseline level with no significant effect on genome integrity (25, 42), it may be suggested that under conditions used in the present study both tested materials showed more than acceptable level of biocompatibility in terms of cyto- and genotoxicity. Thus, observed changes in the level of the primary damage to DNA due to both composite treatments, though some of them showed statistical significance, may be of no biological relevance. Importance of the presented results lies in the attempt to indicate the most suitable procedure for their polymerization and preheating options. For both tested composite materials, the first null-hypothesis was accepted and the second one was rejected. These results with a cell culture cannot be directly used for explanation of the in vivo scenario. However, this does indicate a constant need for finding more advanced procedures and composite resin modifications which would improve polymerization of composite materials and minimize the potential risk for patients and dental personnel.
CONCLUSION
For the Z100, the highest percentage of viable cells was recorded after direct light curing, followed by light curing through CRP and through LDC. For Filtek Supreme Ultra highest percentage of viable cells is recorded while curing through CRP followed by LDC and direct light curing.
For Z100, the highest percentage of viable cells is recorded after preheating on T1 for CRP, LDC and direct light curing. For Filtek Supreme Ultra, highest percentage of viable cells is recorded while preheating composite at T2 for CRP and direct light curing and T1/T3 for LDC.
In the negative control and unpolymerized samples of both tested composite materials, Z100 and Filtek Supreme Ultra, the frequency of apoptotic lymphocytes was slightly higher than the frequency of nectotic lymphocytes. For Filtek Supreme Ultra, apoptosis predominated over necrosis in almost all polymerized samples regardless of the preheating temperature. However, for Z100 composite material apoptosis predominated over necrosis only in samples prepared with direct light curing, regardless of the heating tempaerature.