The aim of the present study was to assess the caries experience of children aged 6 to 7 years old in a socially deprived suburban area of Brazil's Federal District, using the ICDAS II system and to investigate determinants of dental caries. The survey was carried out in six public schools by three calibrated examiners, on a sample of 835 children. ICDAS II codes had to be converted into dmf/DMF components at surface and tooth levels, resulting in unfamiliar caries variables, to enable some meaningful reporting of the findings. The prevalence of dental caries, including enamel and dentinal carious lesions, in primary teeth was 95.6% and in permanent teeth it was 63.7%. Mean values of d(2)mf(2)-t (enamel and dentinal lesions), d(3)mf(3)-t (dentine lesions), D(2)MF(2)-T and D(3)MF(3)-T indices were 6.9 ± 3.8, 3.2 ± 3.4, 1.7 ± 1.6 and 0.2 ± 0.5, respectively. Enamel carious lesions predominated in the dmf-t/s and DMF-T/S indices. Seven-year-old children had statistically significantly more enamel and dentine carious lesions in permanent teeth than 6-year-old children had. Using ICDAS II, the prevalence of dental caries in both dentitions was very high. In both dentitions, the decay component predominated, with hardly any restorations or extractions observed. The new ICDAS II system leads to overvaluation of the seriousness of dental caries experience and made reporting of outcomes cumbersome. Guidelines on analysing data and reporting results should be agreed upon before this system can be used in epidemiological surveys globally.