Knowledge of the biocompatibility of dental materials is important so as to prevent hazardous effects on the surrounding tissues where it is placed. Different bonding agents are used in combination with different resins to provide a possible effective long lasting restoration [1-3].

In an attempt to simplify the bonding technique, manufacturers over the years strived to decrease the number of steps needed to accomplish successful bonding. Twostep-etch and rinse and one step self-etch bonding agents were introduced. They gained rapid popularity because of their relative ease of use [4]. One step adhesives were made both acidic and hydrophilic to blend all adhesive components into one single solution. Available dental bonding agents which are used to bond resins to enamel and dentin differ in composition as well as proportions there-off. It is believed that the chemical formulation determines the success of the adhesive in the clinical situation. However, research done on the biocompatibility of dental bonding agents [5] and the ingredients showed that many of the constituents to be cytotoxic [6]. The cytotoxic effects of monomers have also been reported in several in vitro studies [2,7-16].

The normal composition of bonding agents are: resin monomers, initiators, inhibiters or stabilizers, solvents and sometimes inorganic fillers [17]. These components may be released from the resin systems due to incomeplete polymerization and later due to degradation which resulted in potential adverse clinical effects [14,18,19]. Trubiani [15] supported finding of other researchers [14,20] and conclude that research indicated that resin components such as TEGDMA, HEMA, bis-GMA, UDMA are a likely cause of cellular stress and that they have significant cytotoxic effects. However, HEMA were frequently been used as a wetting agent and therefore competes with water for penetration into the dentine. It also copolymerizes with other resin composite monomers [21]. The release of many monomers were investigated but still limited info on the effect of released quantities of additives such as initiators, inhibitors and stabilizers is known [22]. Ratanasathien [2] suggested that it is possible that synergistic interaction between the multiple components may occur resulting in more cytotoxicity than the individual components would have caused by themselves.

Several of the dental adhesives examined were tested in vitro studies where various established cell-lines were used in vitro. The two primary advantages of this type of testing is the ability to control the environment in the laboratory and the subsequent ability to accurately measure the response to the exposed cells. These tests prove to be rapid, inexpensive, sensitive and a repeatable means of ranking the cytotoxicity of dental materials [23].

Improving the clinical behaviour of adhesives and reducing the risk of cytotoxicity can be achieved by adjusting the (proportion and/or amounts) ingredients in adhesives or to develop new monomers [17] with completely new chemicals.

Therefore, the purpose of this study was to determine the cytotoxicity of quite recent dentine bonding agents from different manufacturers.

A Balb/c 3T3 mouse fibroblast cell line (The National Repository for Biological Materials, Sandringam) was maintained and cultured in standard conditions (37˚C under 5% carbon dioxide and 95% humidity) in Dulbecco’s Modified Eagles Medium (DMEM). The medium was supplemented with 10% fetal bovine serum 1% penicillin, 1% streptomycin and 0.2% fungizone, changed every second day and cells sub-cultured using routine trypsin/EDTA procedures.

For the testing of the cytotoxicity of the bonding agents, an indirect modified model [24] adapted from “the model cavity method” [25] was used (Figure 1).

An example of the construction of such a test bottle is shown (Figure 1). Briefly, the model consisted of a screw top with an 8 mm circular hole in the center which was covered with a permeable membrane. In this way the cap formed an opening (cavity) of 1.5 mm deep, with a surface area of 50.3 mm², which was loaded with 8 µl of bonding agent and light cured according to each manufacturer’s instructions. The sample bottles were then inverted to expose the 400 µl medium to the bonding agent through the membrane and placed in an incubator at 37˚C for 24 hours. Any toxin, if present in the sample to be tested, will then leach out through the membrane barrier into the DMEM. Ten sample bottles were used for

each of the bonding agents.

For cytotoxicity testing, 3T3 cells in the active growth phase displaying 80% - 90% confluence were sub-cultured and diluted to a final cell suspension containing approximately 3 × 10 (5) cells/ml. Cells were plated out in a 96 well plate , 20 wells for each bonding agent (each sample divided into 2 wells) with 10 extra wells to act as controls. On day two the 400 µl of exposed eluate from each of the 10 bottles was used to replace the medium in the 20 wells (200 µl/well). The culture medium of the controls was replaced with fresh DMEM. After 24 hours the widely used MTT colorimetric assay was used to evaluate toxicity. This assay involves the ability of viable cells to use mitochondrial dehydrogenase enzymes to convert MTT (a soluble tetrazolium salt) to a blue/violet formazan end product [26]. 20 µl MTT (5 mg/ml in phosphate-buffered solution) was added to each well and left for a further 3 hours to incubate at 37˚. The medium was discarded to eliminate the MTT and the precipitated formazan crystals were solubilized with 100 µl/well of dimethylsulfoxide (DMSO). Cells with medium not exposed to bonding agent were used as negative controls. Absorbance was measured at wavelength 540 nm on a spectrophotometer to determine the number of viable cells. Every dental material was tested 3 times.

The bonding agents (Table 1) tested were XP Bond, Adper Scotchbond SE, Adper Easy Bond, Gbond, Xeno V, Xeno V+ and iBond.

Only one product Scotchbond SE is a 2-step system (all 6 others being one step systems) with components A and B. In this case the two components were also tested separately.

The Box and Whisker plots (Figure 2) were used to compare distributions resulting from the different survival rates of the 7 bonding agents.

Due to the differences in dispersion the Tukey-Kramer Multiple-Comparison Test was used to compare the locality of the 7 bonding agents. The Box and Whisker plot (Figure 2) shows the percentage viable cells of the different bonding agents. Each Box plot shows the median percentage viable 3T3 cells, 1st quartile value, 2nd quartile value, maximum and minimum values. The region between the 1st and 3rd quartiles shows the location of 50% of the values.

The sample sizes were uni-model but relatively small. Therefore, nonparametric tests were employed and the Kruskal-Wallis test used on the medians to indicate whether differences existed. The Tukey-Kramer Multiple-Comparison Test was used to test for significant differences between pairs of bonding agents.

Table 2 shows the results of the Tukey-Kramer statistical tests based on a 5% significant level. Median values