In vitro cytotoxicity of a composite resin and compomer

[Anonymous]

Discussion

A dentine barrier, simulating the clinical situation, was used to determine whether the two dental restorative materials were cytotoxic in vitro. The dentine disc acted as a physical barrier between the restorative material and the underlying culture medium and cells. However, a dentine thickness of only 70 microns was chosen, because a barrier of increased thickness may mask the material’s potential cytotoxicity. Earlier work by Tyas (1977) using dentine chips showed the toxicity of zinc phosphate cement to be substantially reduced by such a barrier. Although it is not possible to make a direct comparison between two different cell lines, this would appear consistent with the more rapid appearance of toxicity in the HL-60 cells incubated in the absence of the dentine barrier, where both materials resulted in substantial cell death after 12 h, as compared to the adherent ECV-304 cell-line in the presence of the barrier, where substantial toxicity was not observed until after 2 or 3 days of incubation. To mimic the clinical situation more closely, primer was placed on the dentine prior to placement of the restorative material. This layer forms a thin barrier between the restorative and the pulp. However, the effectiveness of this barrier, in terms of limiting ingress from the restorative material, is not known.
Whereas the MTT assay is a good indicator of cell viability, the LDH assay is an indicator of cell lysis. The results of these two assays revealed that, in the presence of the 70 m dentine barrier, Dyract® AP had no effect upon cell lysis until the third day of incubation, whereas the toxic effects of Spectrum® were evident after only 2 days’ incubation. Thus the cytotoxicity exerted through dentine slices by the Dyract® AP restorative material was less than that of Spectrum®, which was extremely cytotoxic and more likely to evoke an adverse pulpal reaction.
During cavity preparation, prior to placement of the restorative material, the mechanical action of the bur may cause microclefts in the dentine. In such cases, the restorative material would be in direct contact with the pulpal tissue, thus magnifying the potential to evoke an adverse pulpal response (Swift et al. 1995). In the present work a suspensory cell-line (HL-60) was used, in the absence of the dentine barrier, for studies on such direct toxicity and on the mechanisms underlying the toxicity of these restorative materials.
Microscopic examination of the stained cells revealed both restorative materials cause cell death, with Spectrum® being more toxic than Dyract® AP. When fully cured materials (40 s light-curing) were used, the predominant form of cell death was apoptosis. In contrast, the main form of cell death following exposure to partially cured materials (4 s light-curing) was necrosis.
Fully cured samples of Spectrum® and Dyract® AP caused activation of caspase-3. Thus, in agreement with the morphological staining data, the fully cured samples induce apoptotic cell death. However, exposure to the partially cured (1 and 4 s) materials resulted in less caspase activation than the fully cured materials. This contrasts with the morphological staining results that showed the partially cured samples to be more toxic than the fully cured material. This would indicate that whilst the partially cured samples cause some apoptosis, the materials provide such an overwhelming insult that the majority of the cells die rapidly by necrosis, and the less cured the material is, the greater the toxicity. Apoptosis and necrosis have been shown to occur together in some other systems. For example, in myocardial infarction cell death is associated with rapid necrosis immediately around the central ischaemic zone, whereas outside the central zone cells die more slowly by apoptosis (Cohen 1993).
Even in fully set restorative materials substantial amounts of short-chain polymers remain unbound, with the result that there is possible elution of leachable toxic components toward the pulp (Ferracane 1994). This may be compounded by inadequate light-curing. In some instances the light-curing unit may emit light that is not at the optimum wavelength (490 nm) for complete setting of the material (Miyazaki et al. 1998). As a result, the restoration might appear solid to probing but would in fact, be incompletely set and capable of initiating adverse pulpal reactions. Poor awareness of the need for regular maintenance of the light intensity of light-curing units amongst dentists has been reported (Martin 1998). In the clinical situation necrotic cell death would result in an inflammatory response and possibly secondary tissue damage to the patient. The present findings indicate that the degree of light-cure is an important factor in the toxic potential of a material, where an inverse relationship exists between the degree of light-curing and restoration cytotoxicity.