Establishing quantitative light-induced fluorescence cut-offs for the detection of occlusal dentine lesions.

The aims of this in vitro study were, first, to define suitable cut-off values for the detection of occlusal dentine lesions by means of quantitative light-induced fluorescence (QLF), and, second, to compare the validity of light microscopy (LM) and microradiography (MR) as reference standards. Fifty-four third molar occlusal fissures with sound sites or non-cavitated lesions were examined. Standard QLF equipment was used to capture, display, store, and analyze fluorescence images. Each tooth was cut into sections in the bucco-lingual direction using a microtome saw, and the caries levels were subsequently assessed with the two reference methods. The histological examination with LM indicated that 46% of the specimens had carious lesions progressing into the dentine compared with 41% measured using MR (D3-4 level). Using optimal cut-off levels for the detection of dentine lesions, the sensitivity and specificity values were > 80%. The area under the receiver operating characteristic curve was slightly higher for MR (0.91-0.93) than for LM (0.88-0.89). The results of this study give a good idea of suitable cut-off points for the detection of occlusal dentine lesions using QLF. Owing to the small number of specimens in this study, further studies should be carried out before establishing definite cut-offs.