Morphometric assessment of enamel demineralisation from photographs

To investigate the validity and reproducibility of a method of morphometric assessment of enamel demineralisation.

An in vitro investigation was carried out on 22 human teeth. One investigator coated the crowns of the teeth with an acid-resistant varnish, leaving a small window on the buccal surface. This was incrementally occluded by varnish over a 14-day period, during which the teeth were placed in a demineralising gel at pH 4.5. After varnish removal, a second investigator blindly quantitated the demineralised area by three methods of examination; direct visual, microscopic and from photographs. The microscopic and photographic measurements were carried out using a morphometric assessment with a 121-dot array. Photographs and assessments were repeated after 1 week. The readings were analysed using the kappa statistic, the limits of agreement and the coefficient of repeatability.

Photographic assessments demonstrated excellent agreement for grid positioning (kappa > 0.81) and substantial agreement for reading reproducibility (kappa = 0.61-0.80). The coefficients of repeatability were found to be the same for repeat readings of the same slide and the repeated slides (5.0 mm2). They were higher for the microscopic technique (6.8 mm2) and for the direct visual technique (7.8 mm2). The limits of agreement are presented graphically.

The photographic technique used was a reproducible method of measuring artificial enamel demineralisation. Measurement from photographs was more reproducible than direct measurement with the naked eye. Subjectiveness of the index leads to most variation and more objective means of assessing enamel demineralisation need to be found.