Diagnostic methods for stained pits and fissures caries lesions on the occlusal surface

ABSTRACT Objective: The aim of this study was to compare methods for in vitro diagnosis of pigmented pits and fissures on the occlusal surface. Methods: Forty-one human molars and premolars extracted with pigmentation in pits and fissures were used. The teeth were evaluated with artificial illumination for classification according to ICDAS, followed by fluorescent laser evaluation. For the radiographic evaluation, the teeth were positioned to obtain interproximal images, adopting the criterion of Rodrigues et al. 2008 for diagnosis of lesions. For histological evaluation, the teeth were sectioned in the mesiodistal direction in the occlusal sulcus region and evaluated in a 20-fold stereoscopic magnifying glass, using the criterion of Ekstrand et al. (1997) for evaluation. The methods were compared with the histological evaluation, obtaining values of accuracy, sensitivity and specificity regarding the response variables “presence / absence” of enamel caries. Results: When comparing visual inspection with histological analysis, 65.85% accuracy, 40% sensitivity and 90.48% specificity were observed with false positive and negative probability of 20% and 38.71%. When comparing the fluorescent laser with the histological analysis, the accuracy was 58.54%, sensitivity 70% and specificity 47.62%. Comparing the radiographic analysis with the histological analysis, it was obtained 53.66% accuracy, sensitivity of 42.11% and specificity of 73.68%. Conclusion: The ROC curve and the Yuden index showed that visual inspection presented the best combination of sensitivity and specificity than fluorescent laser and radiographic analysis for diagnosis, which represented the best method for detecting inactive carious lesions in enamel.