Accuracy of a chair-side test for predicting caries risk compared with established methods. A pilot study.

PURPOSE The aim of this study was to determine the prognostic accuracy of a new chair-side test, determination of lactic-acid production on the tongue, for caries-risk assessment according to caries increment within two years. METHODS 35 children (6-8 years old) participated in the study. The past caries experience (df-t1, DF-T1), caries at second deciduous molars (df-2.mol), approximal plaque index (API) were the clinical, and salivary counts of mutans streptococci (MS)/lactobacilli (LB) and lactic-acid production on the tongue (LAP) the subclinical predictors recorded at baseline. Two years later, the dental examination was repeated (df-t2/DF-T2). According to the increase in DF-T (delta DF-T), caries-risk groups (delta DF-T = 0 versus delta DF-T > or = 0) were built. To assess the prognostic accuracy sensitivity and specificity were calculated. Statistical analysis was performed using Spearman's test for nonparametric correlations. RESULTS The DF-T increased from 0.3 +/- 0.8 to 1.2 +/- 1.6. 20 children were in the no-caries-risk and 15 in the caries-risk group. The p-value for the correlation of delta DF-T to API was 0.007 and to LB counts 0.069. The highest sensitivity and specificity levels were found for LB (65%/70%), LAP reached a maximum sensitivity of 33% with a specificity of 55%. CONCLUSION According to the results, because of its insufficient prognostic accuracy the LAP test is not suitable as a sole screening-tool for an accurate caries-risk assessment.