The experimental production of artificial carious lesions is commonly used in various extra- and intraoral cariogenicity tests aimed at gaining more insight into the mechanisms of caries formation and prevention. These lesions are produced in either bovine or human enamel using various types of demineralization solution under different ambient temperatures (such as room temperature or 37°C) for a particular length of time during which a certain degree of demineralization of the enamel is expected to produce a subsurface lesion with a sound surface layer. In many cases the enamel is coated with layers of acid-resistant nail varnish leaving a window, the position of which depends on the morphology, malformations, or defects of the tooth.

The composition of the various demineralization systems has been developed in an attempt to simulate conditions in the immediate environment of the teeth in vivo during demineralization. The type and concentration of acid used determine the rate of demineralization and the character of the lesion, with acetate buffers producing a deeper lesion, and more quickly than lactate buffers at the same pH and concentration. The inclusion of calcium, phosphate and fluoride in an acidic buffer system partially saturates the solution, resulting in a lesion with well-formed subsurface and surface layers characteristic of the natural lesion.

Many de- and remineralization experiments have been conducted at 37°C as opposed to room temperature, based on the fact that temperature increases the rate of diffusion of chemical species through enamel and aqueous solution, and hence the rate of demineralization of enamel.

Human enamel exhibits site (positional) variation in mineral content and in caries susceptibility. Weatherell et al. (1981) reported a higher content of fluoride towards the occlusal edge of the enamel, while Rolla and Bowen (1977) found a higher content in the cervical region, which they attributed to contact with plaque and gingival fluid that has a high content of fluoride. Poole et al. (1981) showed that, in the same tooth, cervical enamel has lower mineral density than occlusal enamel, while Naujoks et al. (1967) found no significant differences in chemical composition between different parts of human permanent teeth. Many researchers report a variation in the rate and degree of de- and remineralization between different sections and within the same sections of a tooth enamel. For a reasonable comparison to be made between specimens from the same enamel surface, these variations must be considered, as Strang et al. (1987) observed that the extent of remineralization of a lesion depends on its initial demineralization.

Bovine enamel is widely used alongside human enamel in the production of artificial carious lesions, under the same conditions and with the same demineralization system, and it has been established that the human enamel is harder and less porous than bovine, and therefore less rapidly demineralized. This calls for the need to establish the length of time for which the bovine enamel should be exposed to produce an artificial carious lesion as compared to human enamel under the same conditions.

Our aims now were, firstly, to determine the effect of ambient temperature, length of exposure, position on enamel surface and type of demineralization solution on the degree of demineralization of bovine enamel for the production of artificial caries lesions for use in intraoral cariogenicity tests. Secondly, to establish the conditions for the formation of caries-like lesions in bovine enamel, using an acidified gel system and a partially saturated acidic buffer solution at room temperature (approx. 20°C) and/or 37°C.

Forty freshly extracted bovine incisors were polished with wet pumice to remove organic contaminants. A carbon pencil was used to map out three horizontal windows on the incisal, middle, and cervical parts of the labial surface of the enamel of each tooth. Each of the teeth was then cut vertically into two halves with a water-cooled diamond wafering blade (Buehler, Warwick, U.K). Each half was then cut vertically into three parts, and these specimens were air-dried and coated with two layers of acid-resistant nail varnish (Max Factor®) leaving three exposed windows (3×2 mm) on the incisal, middle, and cervical parts of the labial enamel. The hollow cavities (pulp chamber) in the specimens were filled with blue dental inlay wax.

Two types of demineralization solutions were used. (A) Acidified gel system: 0.1 M lactic acid solution was produced, the pH was raised to 4.5 with 0.1 M NaOH, and 6%w/v hydroxyethylcellulose (Aldrich, Dorset, UK) was added slowly with continuous stirring. (B) A partially saturated acidic buffer system: This solution contained 2.2 mM KH₂PO₄; 50 mM acetic acid 2.2 mM of 1 M CaCl₂ and 0.5 parts/10⁶ fluoride; the pH was adjusted to 4.5 with concentrated KOH.

Four groups were formed for experiments to be made at either room temperature (approx. 20°C) or 37°C using either or both of the demineralizing solutions in order to make the following comparisons:

1. gel vs buffer, both at 20°C;

2. gel vs buffer both at 37°C;

3. buffer 20°C vs buffer 37°C;

4. gel 20°C vs gel 37°C.

Each group consisted of two experiments. Ten teeth cut vertically into two halves were used for each group, and one half of each tooth (which was further cut vertically into three parts as described above) was used in each of the two experiments in one group. The three parts of each half were withdrawn from the demineralizing solutions after 3, 4, or 5 days. The specimens that were immersed in gel (20 ml/specimen) were mounted on glass rods with inlay wax; those that were immersed in buffer (20 ml/specimen, continuously stirred) were suspended by a cotton thread.

The specimens were withdrawn after the required length of demineralization, washed with deionized distilled water and air-dried. The nail varnish was removed with acetone, and, using a water-cooled diamond saw (Well, Walter Ebner, Le Locle, Germany), enamel slabs were cut from each specimen from the incisal, middle, and cervical windows, giving a total of 90 slabs from each of the two experiments in one group. One section of approx. 250-?m thickness was cut from each slab, mounted on a brass anvil with nail varnish, and polished with a diamond disc to give a planoparallel specimen of 80-?m thickness.

Each specimen was imbibed with water and examined under polarized light using a Nikon Optiphot® light microscope (Nikon, Tokyo, Japan) with a rotating stage, polarizer and analyser, at a magnification of 450×. This examination was an initial screening for the production of a regular subsurface enamel lesions. Though not all specimens had subsurface lesions, none was eliminated at this stage; all sections were processed for microradiography and image analysis.