A new in vitro model for dentin permeability studies is presented.
The model simulates the positive pulpal pressure of natural teeth, it is suitable for long term studies and requires no sophisticated equipment.
Among all the constituents of the cement, only two compounds were able to diffuse through dentin in detectable concentrations.
The amounts of HEMA that eluted through a 0.85 mm dentin disk under a pulpal pressure of 14.1 cm H 2 O, did not reach toxic levels for the pulp.
To evaluate the elution of HEMA, BPA, UDMA and BisGMA from a conventional resin cement (Multilink Automix ® , Ivoclar Vivadent) through human dentin, under constant positive pulpal pressure.
Ten human dentin disks (n = 10) were adjusted in a new testing device and transparent glass slabs were luted with Multilink Automix ® resin cement, following manufacturer’s instructions, under a steady pressure of 25 N. The device was filled with Ringer’s solution. At 5 min, 20 min, 1 h, 2 h, 21 h, 3 days, 7 days, 10 days and 21 days time intervals, the whole eluate was retrieved from each one of the ten specimens and then, the specimens were refilled with fresh Ringer’s solution. The eluates were analyzed by High Performance Liquid Chromatography (HPLC).
HEMA was detected in the eluate of all of the specimens, from 5 min until 10 days. At four of the specimens, HEMA was also detected in the 21 days eluate at very low concentrations. BPA, UDMA and BisGMA were not detected at any eluate. An unknown compound was also detected at 4.4 min.
The concentrations of HEMA that enabled to diffuse from Multilink Automix ® cement in an aqueous solution, through a dentin barrier, did not reach toxic levels and BPA, UDMA and BisGMA were not detected at all.
Resin cements are an excellent choice for luting indirect dental restorations. Even though they are expensive and technique sensitive, resin cements adhere to the tooth structure and offer enhanced mechanical and physical properties good marginal adaptation, reduced microleakage and solubility , increased fracture resistance of the overlying restoration for some types of all ceramic materials and excellent esthetics. Unfortunately, there are concerns about their biocompatibility.
The organic polymerizable matrix of composite resin materials is a source of compounds that cause a wide variety of adverse effects. At the end of the polymerization reaction, there are always unreacted monomers or oligomers trapped in the highly crosslinked polymer network. The degree of conversion (DC) for most composite resins has been reported to vary from 55% to 70% . The unreacted monomers are susceptible to elution. In fact, almost any component present in a dental resin such as monomers, additives even the fillers, is capable of leaching from the set material . Besides the unreacted monomers, degradation of resins, caused by hydrolysis and/or enzyme catalysis from saliva and enzymes in the oral environment, is another major problem, as it lasts for the entire life of the material. This means that the material leaches monomers continually . By ingestion through the gastro-intestinal track, by diffusion through dentin and by inhalation through lungs, the released compounds can enter the human body and be absorbed by tissues.
During the last two decades several researchers have extensively studied the genotoxic, mutagenic and cytotoxic effects of (di)methacrylates . Monomers show time and concentration depended cytotoxicity, which ranges from 3.6 mM to 10 mM for HEMA, 0.05 mM for UDMA, 0.001 mM for BisGMA according to different cell lines that were studied . Monomers are found to exhibit adverse effects at sublethal concentrations too and are also responsible for the 5–10% of the total cases of allergies from dental materials, the vast majority of which, were caused by HEMA. Only a few cases were due to TEGDMA and BisGMA .
Bisphenol A (BPA) is not a component of dental resins, but it may be present as an impurity in some resins. BPA is used in the synthesis of monomers, such as BisGMA and BisDMA. Elution of BPA may result from impurities left after resin synthesis or from resin degradation. Its estrogenicity is well established . Some researchers found that BPA leached into saliva of treated patients with some commercial composites and sealants while others did not corroborate their results .
In the literature, several in vitro and theoretical models have been used for the study of dentin permeability. Each one of them has its own limitations. In this paper, a new model for dentin permeability studies is presented. As dentin is considered as “a permeable barrier”, compounds derived from dental materials are able to diffuse through dentin into the pulp chamber space and harm the pulp tissue . The factors that affect diffusion are: the applied concentration, the molecular size and the diffusion coefficient of the compound, the thickness of dentin and the area available for the diffusion. The available area for the diffusion, in the case of dentin, is strongly related with the dentinal tubule density and tubule diameter, the presence or the absence of the smear layer and the dentin’s closeness to the pulp . Over more, due to the presence of the positive intrapulpal pressure in natural teeth, when dentin is exposed during clinical procedures, there is an outward movement of the dentinal tubule fluid, which affects the inward diffusion of the compounds . Vongsavan and Matthews demonstrated that the Evans blue dye could not penetrate into exposed dentin in vivo conditions while it could, in vitro experiments and Pashley and Mathews demonstrated a 50–60% reduction in the diffusion of radioactive iodine through etched dentin when the positive pulpal pressure was applied to their model. The positive intrapulpal pressure in human teeth has been estimated at about 14.1 cm H 2 O . The model that is presented in this study simulates the positive pulpal pressure, utilizes a uniform dentin disk thickness and also calculates the amounts of the resin cement and the area of the dentin disk in each specimen.
The primary aim of this study is to investigate the elution of the constituents of a commercial resin cement, Multilink ® Automix cement & primers (Ivoclar Vivadent), through human dentin, under positive pulpal pressure. The compounds HEMA, UDMA and BisGMA, which are constituents of Multilink ® Automix and also the compound BPA, were examined.
Materials and methods
Ten, caries and restoration free, human third molar teeth which were extracted for dental reasons, unrelated to this study, were used. All volunteers (18–30 years old) agreed that after the extraction, their extracted teeth could be used for experimental purposes. The Ethical Committee of the Dental School of the Aristotle University of Thessaloniki approved the experimental purpose of this study. Wisdom teeth were thoroughly cleaned under tap water and then stored in deionized water with 0.02% w/v thymol until the day of the experiment. Only one dentin disk 0.85 ± 0.05 mm, was cut from each tooth, just above the level of the pulp horns, by a low speed saw ISOMET (Buehler, USA) under constant water coolant. The dentin disks were hand-sanded under tap water, by a 600 grit silicon carbide paper to achieve a uniform flat surface with smear layer, acid-etched on both sides, with 35% phosphoric acid (Ultra etch, Ultradent, USA) for 15 s and then thoroughly rinsed with water spray. A custom made glass plate (2.5 cm × 2.5 cm × 3 mm) with a central hole (diameter 3 mm approximately) was attached to a three way stop cock (Polymed, Haryana, India) with cyanoakrylic glue (Super glue, UHU-Bison, PRC) taking care that the three way did not exceed from the glass plate. At the other side of the custom made glass plate, a glass slab (Slab A) with a standard central hole (32 mm 2 ) was attached with aquarium marine silicone to the custom made glass plate and then, the dentin disk was attached above the central hole, with a minimum quantity of aquarium marine silicone (Top sil, Mercola, Athens) placed carefully peripherally, to the enamel margins of the dentin disk. After the aquarium marine silicone was set, all the joins were reinforced externally with sticky wax (Kem-den, Purton,England) and the whole system was filled with Ringer’s solution (Vioser, Greece) to be checked for leakages. If there was a leakage, the system was discarded. The specimen was turned up side down and at the free end of the three way, a calibrated glass tube was fixed. Now, the construction resembled a tooth from the upper jaw ( Fig. 1 ). In order to simulate positive intrapulpal pressure at 14.1 cm H 2 O, the specimens were filled with Ringer’s solution up to 14.1 cm height, counting from the internal surface of the disk and then, the free end of the calibrated glass tube was filled with parafilm. The total volume of Ringer’s solution was 700 μL, including the three-way and the glass tube.
Pictures of the dentin disks were taken by a digital camera (Canon EOS 600D connected with a macro lens, Sigma 105 mm) and the surface area of each disk was calculated by Image J, a license free software program. Then, a 40 μm space for the resin cement was constructed. The distance between the glass slab A and a new glass slab (Slab B), placed on the top of the disk was calculated by a digital caliper. Three sheets of tin foil were place on the top of the dentin disk and four acrylic stops from kalocryl resin (Speiko, Germany) were constructed on the glass slab A around the dentin disk. The distance between glass slab A and glass slab B, was calculated again and ought to be 40 μm greater than the previous measurement. The tin foils were removed, the dentin disk was cleaned with water spray and the specimen was filled with fresh Ringer’s solution.
Following precisely the instructions of the manufacturer, a transparent glass slab (Slab B) was bonded on the dentin disk of the specimen with Multilink ® Automix resin cement (Ivoclar Vivadent, Schaan, Liechtenstein). Multilink ® Automix cement is a conventional self curing resin cement with a photocuring option which is used in conjunction with Multilink ® Primers, a self-etching adhesive system, offered in two bottles. The manufacturer also supplies in the system pack of Multilink ® Automix cement and Monobond Plus, an one component primer which mediates a bond between luting cements and dental materials. The composition of the primers and the cement is shown at Table 1 . As most dual cured cements, Multilink’s ® Automix physical properties differ according to the way of its polymerization. Additional light curing leads to improved properties and Multilink ® Automix in dual cure mode, is found to exhibit % DC equal to 61.3% . For this study, the dual cure mode was selected and the whole procedure is thoroughly described.
|Multilink ® Primer A||Multilink ® Automix||Base||Catalyst|
|Water||85.7||Dimethacrylate and HEMA||30.5||30.2|
|Initiators||14.3||Barium glass filler and silica filler||45.5||45.5|
|Multilink ® Primer B||Catalysts and stabilizers||1.0||1.3|
|Phosphonic acid acrylate||48.1||Pigments||<0.01|
|Methacrylate mod. polyacrylic acid||3.8|
A light film of monobond plus was applied for 60 s on the transparent glass slab B and then dried under a light stream of air. One drop of multilink primer A and one drop of multilink primer B were mixed. The mixture was scrubbed on the dentin disk for 30 s with a microbrush and then the excess was removed by a light stream of air. Multilink ® Automix cement was applied on the dentin disk from the auto mix syringe tip and the transparent glass slab B was placed on the top with light finger pressure at first. Then, a force (25 N) was applied on the glass slab B by a digital force gauge (Chatillon, DFE Series Digital force gauge). Multilink ® Automix cement was cured for 20 s from the lateral side of the specimen. Then, the specimen was removed from the digital force gauge device and the specimen was cured for another 20 s with the tip of the curing device (Power blue Heraeus Kultzer) in contact with the transparent glass slab B.
At 5 min, 20 min, 1 h, 2 h (early elution), 21 h, 3 days, 7 days, 10 days, and 21 days (late elution) time intervals, the whole eluent from each of the ten specimens was retrieved and then the specimens were immediately filled again with fresh Ringer’s solution. All specimens were stored in an oven at 37 °C (Memmert, GmbH Co, Schwabach FRG, Germany).
The analysis of the samples was carried out by HPLC system, which consisted of an automatic Injector SIL-9A, an LC9AD pump by Shimadzu (Kyoto, Japan), a FCV-9A solvent mixing system, a class-M10A computing integrator and a photodiode array detector SPD-M6A set at 210 nm. Degassing of the solvents was carried out using a DGU-2A system. All the examined compounds were obtained from Sigma–Aldrich (Steinheim—Germany), while acetonitrile (CH 3 CN) of HPLC grade and HPLC water were obtained from Fisher Scientific Limited (Bishop, Meadow Road, UK).
A method suitable for the chromatographic separation of five compounds (HEMA, BPA, TEGDMA, UDMA, BisGMA) was developed. The chromatographic separation was achieved within 7.9 min under the following conditions:
Column: Perfectsil target ODS 3 (25 cm × 4.6 mm id, 5 μM) MZ-Analysentechnick GmbH.
Mobile phase: Gradient elution: 45% CH 3 CN/55% H 2 O → 0–2 min; 88% CH 3 CN/12% H 2 O → 2–8 min, followed by 3 min equilibration time.
Flow rate: 1.5 mL/min.
Detection: UV 210 nm.
Aliquots of 100 μL were used for the analysis. Standard stock solutions of 1000 mg/L were prepared in methanol, while working standards, mixtures of the five compounds, in the range 0.5–10 mg/L were prepared in Ringer’s solution. Calibration curves were constructed using peak areas of eluted peaks of HEMA, BPA, TEGDMA, UDMA and BisGMA. A typical chromatogram is illustrated in ( Fig. 2 ). The retention time for HEMA was 3.2 min, for BPA 5.9 min, TEGDMA 6.6 min, UDMA 7.4 min and BisGMA 7.9 min. Linear calibration equations are shown at Table 2 .
|HEMA (CasNo868-77-9)||Υ = 756,996Χ − 36,374||0.9982|
|BPA (CasNo80-05-7)||Υ = 293,826Χ − 176,655||0.9970|
|TEGDMA (CasNo109-16-0)||Υ = 446,519Χ − 81,917||0.9978|
|UDMA (CasNo72869-86-4)||Υ = 42,159Χ − 32,981||0.9922|
|BISGMA (CasNo1565-94-2)||Υ = 80,012Χ − 75,173||0.9911|
Data for the concentration of the HEMA compound were summarized by computing minimum and maximum values, means, standard errors (SE), and standard deviations (SD) per time interval. The effect of time on the HEMA’s concentration was tested by the ANOVA method with repeated measurements (9 time intervals). First, the results of the multivariate approach to repeated measures analysis were evaluated (MANOVA). Following, Mauchly’s test of sphericity was assessed. These results were used mainly for estimating the correct standard errors for testing the statistical significance of the differences between time intervals relative to mean values of HEMA concentration. Greenhouse–Geisser adjustment to the degrees of freedom of the time effect and the corresponding error mean square was adopted, because the sphericity assumption did not hold. Mean concentrations between time intervals were compared with the Least Significant Difference-LSD criterion. Prior to MANOVA, the effect of the fluctuations of the dentin disk thickness and the resin cement amount was tested by the ANCONA method. In all hypothesis testing procedures the significance level was predetermined at p ≤ 0.05. All statistical analyses were performed with the SPSS v.15.0 statistical software (SPSS Inc., Chicago, IL,USA).
From the examined compounds, only HEMA was detected. BPA, UDMA and BisGMA were not detected at any sample of any time interval. An unknown compound was also detected at 4.4 min ( Fig. 3 ).
HEMA was detected from the first sample at 5 min until the 10th day sample, in all of the specimens. In four of the specimens, HEMA was also detected at the 21th day sample at very low concentrations.
ANCOVA revealed that the fluctuation of the dentin disk thickness and the resin cement amount, did not affect the results of this study (p = 0.769 and p = 0.133 respectively). Descriptive statistics for dentin disk thickness and the resin cement amount are given in Table 3 . According to MANOVA results the effect of time was statistically significant (Pillai’s trace = 1, F(8,2) = 2492.74, p < 0.001).
|N||Minimum||Maximum||Mean||Standard error (SE)||Standard deviation (SD)|
|Resin cement amount (mm 3 )||10||2.07||3.01||2.60||0.10||0.33|
|Dentin disk thickness (mm)||10||0.76||0.91||0.85||0.02||0.05|
The mean values of the concentrations of HEMA and the standard error of the measurements in relation to time, is presented in the graph ( Fig. 4 ). Descriptive statistical indices are given in Table 4 . The highest mean concentration was observed at 21 h sample, followed by the mean concentration at 5 min time interval. Between these two time intervals, there was not a statistically significant difference (p > 0.05). The lowest mean concentration was observed at 21 days sample and was statistically significant different with the previous mentioned time intervals (5 min and 21 h time interval). The study was also divided in an early (5 min, 20 min, 1 h, 2 h) and in a late (21 h, 3days, 7 days, 10 days, 21 days) elution study of the examined compounds. Considering the values of the mean concentrations of HEMA, the early elution of HEMA followed a rather linear descending trend ( Fig. 5 ) while the late elution followed an exponential descending trend ( Fig. 6 ).