Validity and reliability of laser-induced fluorescence measurements on carious root surfaces in vitro

The aim was to assess the validity of laser fluorescence (LF), visual inspection and surface texture on root caries lesions in vitro, related to histological lesion depths. Agreement on interdevice, interexaminer and intra-examiner levels as well as the influence of discolouration and surface texture on LF readings were investigated. Calibrated examiners assessed lesion colour and surface texture and performed measurements with 2 LF devices in 3 separate series at intervals of 1 week. Sections (300 mum thick) of 64 out of 93 teeth were obtained and examined under the microscope. Lesion depth was assessed with 2 references: from the delineated borderline of the original exposed root surface (ref. I), or if loss of surface continuity, the absolute lesion depth (ref. II). The correlation (Spearman rho) between LF readings and histological depth was low with values ranging from 0.22 (p > 0.05) to 0.31 (p < 0.05). LF 1 and LF 2 were significantly correlated with discolouration (rho = 0.52 and 0.46, respectively) and with surface texture denoted as hard (rho = 0.34 and 0.33, respectively). A significant correlation between colour and histological depth ref. I (rho = 0.51) and ref. II (rho = 0.56) was found. No significant correlation between surface texture and histological depth was found. The reliability, evaluated as intraclass correlation coefficient was for intra-examiner 0.99, interexaminer 0.97 and interdevice level 0.98. Large differences were found between 2 consecutive measurements, and high measurement errors indicated considerable deviation of individual measurements. Based on these findings, the LF device is not suitable as a diagnostic tool for root caries.