Performance of visual inspection, electrical conductance and laser fluorescence in detecting occlusal caries in vitro.

The aim of this study was to compare visual inspection (VI) and electrical conductance (EC) and laser fluorescence (LF) measurements in detecting occlusal caries. VI was based on fissure discoloration and performed with the naked eye. EC was measured with the ECM device (Lode Diagnostic, Groningen, The Netherlands), and LF was assessed with the DIAGNOdent apparatus (KaVo, Biberach, Germany). In extracted human premolars and molars, clinically sound sites (D0-/D1-lesions), enamel caries (D2-lesions), and dentinal caries (D3-/D4-lesions) were identified using recommended cut-off values. Thereafter, the teeth were cut longitudinally and analyzed by scanning electron microscopy for verification of caries depth. Reproducibility of VI was good, that of EC and LF excellent. In identifying caries at both the enamel and dentin level, the sensitivities of VI and LF were significantly (p < 0.05) higher than that of EC, while EC was significantly (p < 0.05) more specific. The positive predictive values, however, did not exceed 43%. Improved diagnoses at the dentinal level were obtained, when EC and LF were used as an adjunct to VI and when cut-off values were raised. Thus, visual inspection relying exclusively on fissure discoloration seems to allow only proper identification of sound occlusal surfaces. In cases of discolored fissures, the appliance-based methods help to avoid false positive identification of dentinal caries. However, attainable reliabilities of diagnoses do not seem to exceed about 50% to 60%.