Surface layer of enamel caries: features from microradiography

  Features of the surface layer under micro radiography can also be divided into qualitative and quantitative. Qualitatively, SL is shown presenting gray levels higher than that of the underlying body of the lesion and similar to that of inner normal enamel (when the lesion depth is lower than the enamel layer thickness). Quantitatively, it should be expected to present mineral content similar to normal enamel and higher than that of the body of the lesion. These are the most common views in the field of Cariology.

Figure 1. Microradiography of a natural enamel caries with the surface layer presenting gray levels higher than that of the body of the lesion and similar to the inner layers (supposedly normal enamel). 

There are, however, some factors that must be taken into account and that might not confirm the classical interpretation of microradiographic features of the SL. These factors are related to: (1) visual perception of brightness, (2) variations in sample thickness, and (3) unreasonable normalization of mineral content.

1)   Visual perception of brightness

Regarding qualitative features, one must take into account some effects of optical illusion during visual interpretation of gray levels in microradiographic images. The human nervous system adds some interpretation bias at both the periphery  (low-level mechanisms) and central area (high-level mechanisms) of the visual neural tract (Andelson, EW. Perceptual organization and the judgement of brightness. Science, 262:2042-2044, 1993). Low-levels mechanisms include the illusion that a gray area is perceived brighter when adjacent to a dark background, and darker when adjacent to a bright background. High-level mechanisms include optical illusions related to luminance and geometry of adjacent gray areas. The perception that the surface layer is similar to normal enamel might be a result of optical illusion.

2)   Variations in sample thickness

        Radiopacity of enamel is directly proportional to its thickness. Mineral content from microradiography is primarily based on gray levels and sample thickness. The same gray level results in a high mineral content where thickness is low, and in a low mineral volume where thickness is high. The point here is that undemineralized sections of enamel caries submitted to microradiography commonly present intra-sample variations in thickness. Heterogeneous sample thickness might result from either diamond wheel sawing, diamond wire sawing, Exackt sawing, or grinding with lapping jips. Figure 2 shows the edge of a ground section of natural enamel caries cut parallel to the prisms paths. Marked variations in sample thickness can be seen from the surface layer inward. This a result of the fact that in a heterogeneous material as enamel areas with different hardness resist differently to the mechanical forces applied during sample preparation. Unaware of that, scientists commonly determine sample thickness from what is measured at the surface or else from what is measured using a measuring device (like a gauge micrometer) for the whole section. The implication is that gray levels are interpreted, for both qualitative and quantitative purposes, on the basis of wrong thickness values. Misguided by the unreasonable assumption that sample thickness is homogeneous across the SL, body of the lesion, and inner enamel, scientists do an incorrect interpretation of the histological features. When variations in thickness exist and are ignored, the resulting misinterpretation might result in a SL with a higher mineral content than the body of the lesion and similar to normal enamel. This is not to say that it always occurs, but that variations in sample thickness must be taken into account. This is currently underestimated, mostly ignored actually.

Figure 2. Photomicrograph of the cut edge of a natural enamel caries showing variations in sample thickness. SL is at the left upper corner, followed by the body of the lesion (dark area). 

3)   Unreasonable normalization of mineral volume

This factor is closely related to the later. For unreasonable normalization, it is meant the normalization of inner enamel mineral content either as 100% mineralized or with any particular per cent mineral content (eg.: 88%). This is commonly seen in plot profiles of mineral content across an enamel caries lesion presenting normal enamel in the inner layers. The profile presents a flat line with a high mineral content in inner enamel (examples can be seen in Arends & Christoffersen, J Dent Res, 65: 2-11, 1986; DOI: 10.1177/00220345860650010201). Experimental data on mineral volume of enamel using the Angmar equation (Angmar et al., J Ultrastruc Res, 8: 12-23, 1963) have never shown flat mineral profile as feature of normal enamel. In fact, it is known that mineral volume decreases from the occlusal to the cervical third of the crown, and from outer to inner enamel (Theuns et al., Arch Oral Biol, 28: 797-803, 1983). The flat profiles are reported in articles that did not measure mineral volume with Angmar equation. They usually simply plot gray levels across the lesion and assume that sample thickness is homogeneous. As mineral content is not measured, but determined by the authors that they represent normal enamel and lower values represent caries, this can be considered as unreasonable normalization. The main reasons for being unreasonable are: rigorous experimental data do not confirm flat profile as a feature of normal enamel (in fact it is expected to decrease from the surface inward), and possible variations in sample thickness are not checked.

      In this context, it is important to note that the presence of a relatively normal SL is not a universal feature of enamel caries (Medeiros et al., J Microsc., 246: 177-189, 2012; doi: 10.1111/j.1365-2818.2012.03609.x). There are reports of mineral content profiles presenting an increasing slope, an indication that mineral content increases from the SL inward (Sousa et al., Caries Res, 47: 183-192, 2013; DOI: 10.1159/000345378 and cited references). Such lesions do not present a clear transition from SL to subsurface, so that the thickness of the SL cannot be measured according to was recently proposed by Cochrane et al. (J Dent Res, 91:185-191, 2012; DOI 10.1177/0022034511429570).