Abstract
The aim of this in vitro study was to assess the diagnostic performance of five methods of using the Electrical Caries Monitor (ECM) to detect caries lesions at enamel and dentine thresholds in occlusal molar surfaces with and without stain. Histological validation was performed in 152 occlusal sites in permanent molar teeth. For the sites without stain the areas under the ROC curve (AUC) for the five methods ranged from 0.69 to 0.93 (enamel) and 0.74 to 0.92 (dentine). For stained sites the AUC ranged from 0.59 to 0.65 (enamel) and 0.66 to 0.77 (dentine). The performance of the ECM was reduced in the presence of stain. The measurements using toothpaste as a conducting medium were generally worse than when direct measurements were taken. The profiles of the resistance curves generated as the tooth was dried for 5 s were used to derive additional summary variables. The AUC for all sites using an outcome variable generated from these data were 0.89 (enamel) and 0.91 (dentine), respectively, for all teeth. It is concluded that the performance of the ECM might be improved by either considering a combination of variables or using different resistance cut-offs in the presence or absence of stain.