Objectives: To test the ability of a site-specific method of evaluating mutans streptococci (MS) present in plaque in order to predict caries risk in children and to compare the results with those obtained from estimating the levels of MS present in saliva.
Methods: Caries prevalence was examined twice, 6 months apart, in 129 preschool children ranging in age from 1.5 to 5. A commercial strip method was used to evaluate the MS score (0-3) in plaque collected from four selected tooth surfaces at two interdental spaces and in saliva.
Results: The screening criteria of the MS score for caries incidence between scores 1 and 2, and scores 0 and 1 were the most valid for evaluating plaque (accuracy of prediction, 80%) and saliva (accuracy of prediction, 71%), respectively. Based on these screening criteria, the evaluation of saliva MS was invalid when the plaque MS score was considered. Screening plaque MS was effective, regardless of caries experience at baseline, according to the chi-square test (p<0.01), but screening saliva MS was ineffective when caries experience was present at baseline. A multivariate logistic regression analysis including six variables showed that the plaque MS score and caries experience at baseline were significantly correlated with caries incidence: caries experience at baseline (OR=5.02, 95%CI: 1.81-14.59) and high plaque MS score (2 or 3) (OR=12.59, 95%CI: 3.18-67.08).
Conclusions: These results suggest that the semi-quantitative evaluation of plaque MS sampled from only two proximal sites on the molars is more effective for screening preschool children at a high risk of developing caries in their primary dentition than other important caries risk indicators.