Objectives: To explore a possible relationship between the caries experience and pattern in the primary dentition at 5 years of age and the permanent dentition at 10 years of age. Further, to examine the possibility of predicting children in a caries-risk group at 5 years verified at 10 years of age.
Materials and methods: A sample of 186 children (90 males) were clinically examined as 5-year-olds and re-examined as 10-year-olds by calibrated dentists. A five-graded diagnostic system including enamel caries was used. Bitewing radiographs were taken. A true risk group of children at 10 years were defined as those with at least one dentin or filled lesion on the mesial surface of 6-year molars, and/or on incisors, and/or total DMFS (decayed, missing, and filled surfaces) more than 1 SD above the mean. The prediction was measured in terms of OR (odds ratio), sensitivity/specificity, and receiver operating characteristic curves.
Results: Statistically significant correlations (r=0.5) were found between the caries experience in the two dentitions as well as between the primary second molars at baseline and the permanent teeth at 10 years. 'Primary second molars' and 'all primary molars' were the most powerful predictors for allocation into the risk group (24% of the sample). The highest achieved sum of sensitivity and specificity, 148%, was attained at a cut-off point above two carious surfaces in enamel and/or dentin in primary second molars.
Conclusions: Statistically significant relationship in disease between the dentitions was found. More than two surfaces with caries experience in primary second molars are suggested as a clinically useful predictor at 5 years of age for being at high risk at age 10.